# Jobably Docs — every page, for language models > Product documentation for Jobably, the job-search platform built for your own AI agent. Short index: https://docs.jobably.com/llms.txt · Platform overview: https://jobably.com/llms-full.txt ## Jobably documentation https://docs.jobably.com/ > The job search that runs itself — scans, scores, tailors, tracks. Upload a resume, answer the questionnaire, create your first search — the first scan queues immediately. [Start here](/getting-started/onboarding/). A few simple connectors find exactly the right roles. Learn the syntax — and the traps Jobably catches for you. [Writing queries](/searches/query-syntax/). Every posting is scored 0–10 against a rubric built from your answers — and your ratings tune it over time. [How scoring works](/scoring/rubric/). Tailored resumes can never invent a figure. How the guard works and what it means for you. [Resumes](/resumes/provenance/). From matched to offer: a kanban board that shows how each application is aging and a full history on every card. [Your pipeline](/board/workflow/). Your matches one card at a time, on your phone. Right to keep, left to pass, and tune your searches on the spot. [The phone app](/app/swipe/). Tailor your resume with Jobably beside you — line by line, role by role, every proposal with its reason. [The studio](/resumes/studio/). Connect ChatGPT, Claude, Codex, Copilot or any MCP client — hand it your resume and it sets up your search. Your daily short list, tailored drafts, applications and reports, 64 tools with cost classes and confirmation gates. [Connect](/mcp/connect/) · [Your agent's workspace](/mcp/workspace/). ## Your first scan https://docs.jobably.com/getting-started/first-scan/ > What happens after you create a search, and how to read scan health. When a search runs, Jobably: 1. **Collects** new postings from each source since the last check (the first run looks back a full week — career-site boards move slower than job boards, and a day-wide first window can miss everything; after that, checks slightly overlap so nothing slips through between runs). 2. **Removes duplicates** — including the sneaky kinds: the same role reposted under a fresh listing so it looks new, and the same role posted under two company names, such as a parent and its subsidiary (Merck and MSD) or a regional arm (Ricoh and Ricoh USA). Two postings count as one only when the title and the location match too, so a real second opening is never hidden. 3. **Filters out** employers on your reputation-exclusion list before anything is saved. 4. **Saves and scores** everything that remains against your personal rubric, each with a written explanation. ## Reading scan health The dashboard shows health for every search: when it last ran, how many postings came back, how many were kept, and any problems — named plainly, not hidden. **Zero results?** The health panel shows exactly how your query was understood. A query that came out too narrow (or empty) is the most common cause — see [the traps Jobably catches](/searches/query-syntax/#traps-jobably-catches-for-you). ## How often Searches run every 6 hours by default — configurable per search, down to hourly on the Executive plan. Pausing a search stops the runs but **remembers exactly where it left off** — resuming picks up from that point with no gap and no re-fetching. That's also why searches pause rather than delete. ## Onboarding https://docs.jobably.com/getting-started/onboarding/ > Resume upload, targeting questionnaire, and your first search. Sign in at [app.jobably.com](https://app.jobably.com) with Google, Microsoft, or an emailed one-time code (if a code doesn't arrive, hit **Resend** on the same screen — the newest code is the one that works, and check spam). Fresh accounts land in a short wizard: upload your resume, then either **quick start** (one click) or the longer form or chat. Using ChatGPT, Claude or another AI agent? It can do all of this for you — see [Let your agent set you up](/mcp/connect/#new-to-jobably-let-your-agent-set-you-up). ## 1. Upload your resume Drop a **docx, pdf, md, or txt** file. Jobably parses it into a structured profile — name, headline, contact, sections — and shows you the parse for review before anything else happens, including the **full parsed text** in an expandable view so you can confirm nothing was lost in translation. This parsed profile is the raw material for tailoring. Every tailored resume is built only from what's actually in it (plus your accomplishment bank), so the better this document, the better every artifact downstream. You can skip this step and upload later from the Resumes tab. ## Quick start: one click from your resume Once your resume is parsed, Jobably proposes a profile and a first search from it — your **current titles**, the **level** they sit at, and **where your current role is based** — and says where it read each one. Add the two things a resume can't tell us if you like (remote, hybrid or on-site, and a salary floor), then choose **Start my search**. Your profile is saved, your first search is created and its first scan runs at once; your first matches arrive within minutes, scored against your profile. Nothing in the quick start is final: change any of it later on the Profile and Searches pages. If you'd rather say more up front, take one of the longer paths below — the same details in more depth. ## Or choose a longer path: form or AI chat Pick how to answer: the classic **form** below, or a **chat with the AI interviewer** — plain-language questions, one at a time, with "skip" always accepted. Both paths gather exactly the same details, and the chat only offers "Create my profile" once every topic is covered (21 in all). You can switch from chat to form at any point and keep everything you've answered. ## 2. Set your bar (the questionnaire) The questionnaire is **the single authority on targeting**. Everything derives from it — the scoring rubric, what surfaces, what gets flagged: - **Seniority floor** — roles a full rung below score at most 4. The ladder covers every function, from senior individual contributor through C-level — not just engineering. - **Preferred locations** — a *ranking* signal, never a filter. A strong fit in the wrong city still surfaces, with a location note in the rationale. - **Work arrangement** — pick as many as apply (e.g. remote *and* hybrid). Again: preference, not a filter. - **Compensation floor** — postings that state less are marked; silence is treated as unknown, not failure. - **Industries** — both the ones you prefer and the ones to **avoid**; an avoided industry is a strong mark against, never a silent filter. - **Title keywords you want** — titles that should count strongly in your favor, with OR/AND if you like (`"VP of Engineering" OR "Head of Platform"`). - **Excluded title keywords** — titles that disqualify regardless of employer (e.g. "sales engineer" when you want platform work). Saving builds a versioned rubric. You can read the exact rubric text on the Profile page — what you see is literally what the scorer is told. ## 3. The interview (optional, worth it) Ten open questions, one at a time, all skippable: your story, what a great next role looks like, non-negotiables, dealbreakers (including the tell in roles that only *look* right), **what your resume undersells**, **what you never want to do again**, company preferences, and any location or compensation nuance the checkboxes can't capture. Skip anything you don't have a view on — a blank answer is more useful than a guessed one. Your answers go into your scoring rubric **word-for-word** — the scorer reads them as first-hand requirements, so plain and specific beats polished ("Remote or Charlotte. Base over $180k. No pre-Series B startups."). It's the single biggest thing you can do to sharpen your match scores, and every answer stays editable on the Profile page. ## 4. Create your first search Name it, build the query with the [guided builder](/searches/query-syntax/) (or write it by hand in Advanced mode), pick sources, and finish. The wizard queues the first scan immediately; a **progress card on the dashboard** shows it running and updates by itself when the first matches land — usually within minutes. After that the search runs on its own cadence (every 6 hours by default; hourly on Executive). Along the way you'll also get a **welcome email** when you join and a **what-happens-next email** once onboarding completes, so the autopilot part is never a mystery. ## Every button, explained https://docs.jobably.com/app/every-button/ > What each action in the app actually does — including what it costs. Nothing in Jobably spends money or changes your record silently. Here is what every action does, screen by screen. ## Getting started | Button | What it does | | --- | --- | | **Upload and parse** | Reads your resume and shows you the parse to check. Free. | | **Start my search** | The quick start: saves the profile Jobably proposed from your resume (plus your answers on arrangement and salary, if you gave them), creates your first search and runs its first scan straight away. Your first matches arrive within minutes and are scored. Everything stays editable on the Profile and Searches pages. | | **Chat with the AI interviewer** / **Answer a quick form** | The longer paths: the same details in more depth, ending in the same place. | ## On a match | Button | What it does | | --- | --- | | **Pursue** | Moves the posting onto your Board as *Pursuing*. Free; reversible from the Board. | | **Tailor resume** | Opens the [tailoring studio](/resumes/studio/) for this posting: Jobably reads the job, walks your resume line by line and role by role, and proposes rewordings and additions from your accomplishment bank with the reason each one serves this posting. You accept, reword or keep every line, then build the document — a resume aimed at this job **plus a matching cover letter**, both as .docx. Every figure is checked against your master resume and bank, and every line must restate something already in your record. One tailoring credit per session, however long the review runs — the build and any revision are included. | | **Prep packet** | Builds an interview preparation brief: company research, likely interview themes, STAR stories from your verified history, and outreach *drafts* (Jobably never sends anything on your behalf). Requires your own Anthropic key under Settings — prep uses live web search, which runs on your key. Uses one prep credit. | | **Re-score** | Scores (or re-scores) just this posting against your current rubric — useful after you've tuned the questionnaire. Live progress on the card; the fresh score replaces the badge the moment it's done. Uses one scoring credit. | | **×** (on a finished status) | Dismisses a completed or failed operation notice so it doesn't linger. Gone for good — it won't come back on reload. | | **Dismiss** | Removes the posting with a required reason and note. The note is the important part: it tunes future scoring, so "wrong level — this is IC work with a VP title" makes tomorrow's scores smarter. | | **Try again** | Appears only when an operation failed for a reason a retry can fix, such as a provider timeout. Nothing wrong with your content. | | **Load more** | Fetches the next page of matches, 50 at a time. | | *Opened from a link* | A link to one job — from your agent, the short list, or the search box — opens Matches with that job **first, expanded and scrolled into view**, whatever your filters or how far down the list it would sit. Dismissing or pursuing it ends the link and brings up the next job. | | **Restore to Matches** | Shown instead of Pursue and Dismiss on a job you passed on. Puts it back as a live match. Free. | | **Close** | On a job you passed on that a link opened: puts it away and brings up the next job. Changes nothing on your account. | | *Job title click* | Expands details: stated pay, key skills, role and requirements summaries, the original posting link, and your package downloads once tailored. | | *Score click* | Opens the **why this score** panel: the written rationale, what the scorer found on seniority, location and pay, and links to the settings that produced it — your interview answers, targeting and exclusion lists, the search's threshold, your Matches floor. A **Re-score** button sits in the same panel for after you change one. Works on Matches, the Board, a card's drawer, and the Dashboard. | ### After a tailor | What you see | What it means | | --- | --- | | **Gaps** panel | What this posting asked for that your record does not evidence. The draft never claims these. Each is either something to add to your [accomplishment bank](/resumes/accomplishment-bank/), something for a cover letter, or a reason not to apply. | | **Package downloads** | Your tailored resume and cover letter. Downloading asks you to confirm you will read the draft first — see [Download](#on-resumes) below. | | **"Built on the included model"** | The tailor ran without your own API key. Slower, and long resumes come back partly untailored. One field in Settings fixes it. | ### Filtering the list | Control | What it does | | --- | --- | | **Status** | Narrows to one pipeline column. | | **Minimum score** | Hides everything below the score you pick. | | **Source** | Career sites, job boards, or postings you saved with the extension. | | **Found by search** | Narrows to one saved search. Archived searches are not listed — their matches stay, only this way of slicing them goes. | | **Found within** | Narrows by how recently the posting arrived. | ## On the dashboard | Control | What it does | | --- | --- | | **Searching now** strip | What is scanning at this moment, when the last sweep ran, when the next one is due, and how many postings today's sweeps checked and how many were new. Keeps itself current without a reload. | | **Recent sweeps** feed | The last sweeps, search by search, with what each found. | | **Score bars** | Click a band to open Matches filtered to that score range. | | *Score click* | The same why-this-score panel as on Matches. | ## On the board | Action | What it does | | --- | --- | | **Move to…** | Advances the card to the next stage (Screen, Interview, Offer…). Only legal next steps are offered; the server refuses illegal jumps, so your history always makes sense. | | **ended: rejected / withdrawn / accepted** | On any card from Applied on. Records how the application ended — their decision, or yours — and moves it to the Closed list with the stage it reached. | | **Drag & drop** | Same as Move to…, with the mouse. Keyboard works too. | | **I applied to this elsewhere** | Records an application you made through a recruiter, a referral, or a company site. Creates the card if the role isn't tracked yet. Declaring the same role twice never duplicates it. | | **Notes** | Free text against a posting — a recruiter conversation, a referral, why it mattered. Notes are what you read later, when the score has stopped being the interesting part. | | *Card click* | Opens the full drawer: description, rationale, downloads, notes, status timeline. | ## On resumes | Button | What it does | | --- | --- | | **Upload** | Adds a new master resume version. Previous versions are kept. Your name, email, phone and every employer, title and date are read from it and used verbatim in tailored drafts. | | **Download** (master) | Retrieves any master version. This is your own file — nothing to review. | | **Download** (tailored) | Asks you to confirm you will read the draft before sending it, every time. That is a step, not a formality: the guards stop a draft inventing a figure, they cannot stop it being aimed at the wrong thing or carrying a licence from another industry onto this application. | | **Pin for Studio** | The Studio starts from this master version instead of your newest upload, and the page warns while a newer master exists. It is not a style reference: for that, mark a final you actually sent as an exemplar under Drafts. | | **Diff vs vN** | Line-by-line comparison between two master versions. | | **Delete** (tailored) | Permanently removes a tailored artifact — asks you to confirm first. The posting, its score and its history are untouched. | ## On searches | Button | What it does | | --- | --- | | **New search** | Creates another standing scan. | | **Run now** | Queues an immediate scan instead of waiting for the search's scheduled cycle. Fetched postings count against your monthly jobs quota. Paused searches must be resumed first. | | **Pause / Resume** | Stops or restarts scanning. The scan watermark is preserved, so resuming never re-fetches what you already have. | | **Edit** | Changes the query, filters, threshold, or cadence, with a live compiled-query preview. Filters you can't see on the form are carried through untouched. | | **Archive** | Hides a search from the working list and pauses it. The watermark survives, and **Restore** brings it back — paused. Searches are never deleted. | ### Suggested searches Built from your questionnaire, shown above your saved searches. | Button | What it does | | --- | --- | | **Create** | Saves the suggestion as a standing search and queues its first scan. | | **Customize** | Opens it in the new-search form so you can adjust before saving. | | **Dismiss** | Hides that suggestion. Remembered on this browser. | ## On profile | Control | What it does | | --- | --- | | **Targeting questionnaire** | The single source of truth for scoring. Saving it re-derives your rubric. | | **Your interview** | Your answers in your own words. They feed the rubric verbatim, and are editable any time. | | **Engagement types** | Salaried, contract, or both. | | **Excluded employers** | Two lists, and the difference matters: *poor fit* is about match quality, *reputation* is a veto that applies whatever the role. | | **Accomplishment bank** | The record tailoring draws on beside your master resume. One accomplishment per line, with real figures — tailored drafts may only use figures that appear here or in your master. | | **Import from LinkedIn** | Upload the data export LinkedIn emails you and every role description becomes an accomplishment. Nothing is replaced; entries are added, duplicates skipped, and you're told what came in. See [your accomplishment bank](/resumes/accomplishment-bank/). | | **Re-score after saving** | Offered when your rubric changes, so existing matches reflect the new one. | ## On settings | Control | What it does | | --- | --- | | **Notifications** | Two groups. **Search & matches**: your own **Matches score floor** (one number for every search, or leave each search's threshold in charge) and how often match emails arrive — as each scoring run lands, once a day, or off. **Summary emails**: the pipeline summary daily, weekly (Mondays), or off. Saved as you change them. The **email alert level** (default 8) decides which matches are emailed and is a different number from any threshold — see [thresholds](/scoring/threshold/). | | **Bring your own model keys** | Store a provider key (write-only — it is never shown again) and pick the tailoring model. Markedly faster than the included model, and it degrades far fewer sections. | | **Tailored resume length** | How many experience bullets a tailored draft keeps. The default goes by role (up to 12 for your current role, 6 for recent roles, fewer for older ones); you can add an overall limit or keep every bullet. Each role keeps at least one; what is left out is listed beside the draft. The same budget applies when a studio session opens, where each bullet left out is marked and can be put back. | | **Personal access tokens** | Create scoped tokens for the browser extension and AI agents, with a ready-to-paste connect command. Revoking keeps the row, dimmed, so you can see what existed. | | **Linked sign-ins** | Connect or disconnect Google and Microsoft. The provider you currently sign in with cannot be removed. | | **Theme** | Light, dark, or follow your system. Remembered across every Jobably surface. | | **Export data** | Everything Jobably holds about you, as a file. | | **Reset account** | Clears matches, scores, artifacts, searches, your master resume and questionnaire — and keeps your login, subscription and tokens. Requires typing `RESET`. | | **Delete account** | Permanent. Requires typed confirmation. | ## On the phone — Jobably Swipe [Jobably Swipe](/app/swipe/) is the discovery app at swipe.jobably.com. | Control | What it does | | --- | --- | | **♥** / swipe right | Keeps the posting: it stays under **Matched** on your board, even one that scored below your floor, and is recorded as a good fit. | | **✕** / swipe left | A sheet asks why; tap a reason and the posting is dismissed with it. **Never mind, keep looking** puts the card back. | | *Card tap* | Expands the card to the full description. **Collapse** brings it back. | | **Undo** | Puts the last card back and reverses what it recorded. | | **Install** (banner or Account tab) | Adds the app to your home screen. Where the browser offers no prompt, the Account tab shows the steps for your phone. | | **Edit / Pause / Resume / Run now** (Searches tab) | The same actions as the app's Searches page; filters the phone form does not show are carried through untouched. | ## Everywhere | Control | What it does | | --- | --- | | **?** (header) | Context-sensitive help for the screen you're on. | | **Sun/moon** | Light and dark theme, remembered across all Jobably surfaces. | | **Avatar menu** | Profile, Billing, Settings, and sign out. | | **Speech bubble** (bottom right) | Sends feedback — bug, idea, issue, or request — straight to us, with the page you were on attached. | | **Corner notices** | Confirmations and failures appear briefly in the bottom-right corner and fade. Errors stay longer and can be dismissed; hovering holds one open while you read it. | ## How it all fits together https://docs.jobably.com/app/how-it-fits-together/ > What feeds what — searches fetch, your profile scores, and where each edit takes effect. Jobably has two engines, and knowing which one an edit feeds answers most "where do I change this?" questions: - **Searches decide what gets FOUND.** Each saved search queries the job boards on its schedule with its own keywords, location, and filters. - **Your profile decides how findings are SCORED.** The questionnaire, your interview answers, and your excluded lists are compiled into your scoring rubric; every posting a search finds is scored against it. ``` Searches ──find──▶ Matches ◀──score── Profile (rubric) │ │ ▲ │ keywords, │ pursue/tailor │ targeting, interview, │ location, ▼ │ excluded industries… │ cadence Board │ └── Searches page └── Board page └── Profile page ``` ## Where each edit takes effect | You change… | It affects… | It does NOT affect… | | --- | --- | --- | | A search (keywords, location, filters, cadence) | What future scans fetch | Scores of anything already found | | Profile targeting or interview answers | Your rubric — how postings are scored from now on | Existing scores, until you **re-score** | | Excluded employers (poor fit) | Ranking signal in the rubric | What gets fetched | | Excluded employers (reputation veto) | Postings from them never surface at all | — | | Score threshold on a search | Which of that search's matches show by default | What gets fetched, or stored scores | | Master resume | Tailoring and future scoring context | Existing tailored resumes | ## The two moments things update 1. **Fetch time** — each search runs on its cadence (and immediately when created or "Run now"). New postings are de-duplicated, filtered, stored, and scored against your **current** rubric. 2. **Re-score time** — when you change your profile, existing matches keep their old scores until a re-score runs. After saving profile changes, the page offers **Re-score my matches** one click away; it's also on the Matches page. ## Common questions this answers - **"I updated my profile — do my searches change?"** No. Searches fetch the same things; the *scores* of new findings reflect your updated rubric, and existing findings update when you re-score. - **"I want different jobs found."** Edit the search (or add one) on the Searches page — keywords and filters live there. - **"I updated my profile but the scores look the same."** They will until a re-score runs — use the offer that appears after saving, or Re-score on Matches. - **"Where do Matches end and the Board begin?"** Same jobs, two views: Matches is the review queue (filter, dismiss, pursue, tailor), the Board is the pipeline you're actively working. Moving a card on either side is the same underlying status change — and before you apply, every move is reversible. ## Jobably Swipe (phone app) https://docs.jobably.com/app/swipe/ > A small installable app for reviewing your matches one card at a time — right to keep, left to pass — and tuning the searches that feed it. **Jobably Swipe** lives at [swipe.jobably.com](https://swipe.jobably.com) and installs on your phone like an app. It signs in with your Jobably account — email code, Google, or Microsoft — and does one job: discovery. It is the app to open for ten minutes on the train. ## Installing it Open [swipe.jobably.com](https://swipe.jobably.com) in your phone's browser and sign in. Where the browser offers to install, a banner at the top of every screen says **Install Jobably Swipe** — one tap adds it to your home screen and it opens full-screen from then on. The same button sits on the **Account** tab. Where the browser does not offer it, the Account tab shows the steps for your phone: | On | Do this | | --- | --- | | **iPhone / iPad** (Safari) | Tap **Share** (the square with the arrow), then **Add to Home Screen**. Safari never shows an install prompt, so this is always the path on iOS. | | **Android** (Chrome) | Open the **⋮** menu and tap **Install app** (or **Add to Home screen**). | | **Desktop** (Chrome or Edge) | Click the install icon at the right end of the address bar, or open the browser menu → **Cast, save, and share** → **Install page as app**. | Once installed, the app opens instantly even with a poor connection — only the shell is stored on your phone. Cards, swipes and search changes always go to Jobably live, so what you see is never stale. ## The deck Every posting your searches found that you have not judged yet is dealt as a card — including postings that scored **below** a search's threshold, marked so you know. Each card reads like a profile: company (industry, size, where it was found), role (title, location, remote / hybrid / on-site, seniority), the posting's stated pay, the score with Jobably's written rationale, the skills and must-haves, and the full description on tap. The **✕** and **♥** buttons are always on screen under the card, whatever the card's length. The first time the deck opens in a session, the top card nudges right with a **Keep** stamp and left with a **Pass** stamp, so the gesture shows itself; touching the card ends the demonstration, and it never plays for anyone who has asked their phone for reduced motion. - **Swipe right (or ♥) to keep.** The posting stays under **Matched** on your board — even a below-threshold one, because a keep is your call and outranks the floor. It is also recorded as a good fit, which tunes future scores. - **Swipe left (or ✕) to pass.** A sheet slides over the card asking why — wrong level, wrong location, comp, industry, company, or just not for you. Tap one and the posting is dismissed; the reason tunes future scores exactly as a dismissal in the app does. **Never mind, keep looking** puts the card back. - **Tap the card** to read everything, including the full description. - **Undo** (top right) puts the last card back and reverses what it recorded. A swipe leaves the screen the moment you let go and is saved behind it. If saving fails, the card comes back and says why — a swipe never silently fails to stick. ## Searches The Searches tab lists the same searches the app runs, each with its status (live, paused, or scanning right now), how many postings it has found, when it last ran and when it runs next. From your phone you can change a search's name, title query, location, remote-only, score threshold and cadence, pause or resume it, and run it now. Every filter you cannot see here is carried through unchanged. Building a new search with the query helper stays in [the app](https://app.jobably.com/searches). ## Account Your signed-in address, the install button or steps, the theme toggle (shared with every Jobably surface), sign out, and a link to the full app. ## What it leaves out Tailoring, the board, interview prep, your profile and rubric — all of that is in the full app, on purpose. Swipe is for clearing the deck: keep what deserves a look, pass on what does not, nudge a search's threshold, and let the pipeline do the rest. ## A tour of the app https://docs.jobably.com/app/using-the-app/ > What each page of app.jobably.com does, screen by screen. Everything in Jobably can be driven from the web app at [app.jobably.com](https://app.jobably.com) — no command line required. This page walks the portal screen by screen; the next page explains [every button](/app/every-button/) in detail. ## Dashboard Your search at a glance: - **New matches** since your last visit, with score chips. - **Score distribution** — a histogram of how postings score against your bar. - **Pipeline** — counts per stage, mirroring the Board. - **Where they go** — the funnel: postings seen → scored → matched at your floor → resume drafted → applied → screen or beyond → closed, each step as a share of the one above (Closed against Applied, since an application can close from any stage). - **Hit rate by search** — the share of each search's postings that reached your floor: the free instrument for "is it the market, or my query". A search shows once it has eight scored postings. - **What it has cost this cycle** — credits spent, per match found and per application sent, from the same reading your plan's meters use. - **What came back** — still open, rejected, withdrawn, accepted, the average days to hear back, and the share answered at all. - **Waiting on a reply** — applications past five days with nothing back. - **Scan health** — every search with its last runs per source, a running **total of matches found**, and plain-language results: "latest scan: nothing new" after a repeat run is healthy — everything returned was already known and de-duplicated. - **Usage this cycle** — meters against your monthly plan quotas. Brand-new accounts also get a **first-scan progress card** here that watches the initial scan and flips to "fresh matches are below" on its own. When setup is incomplete (no resume yet, no search yet), a **Step 1 — finish setting up** checklist appears with direct links, plus a way to restart the guided onboarding at any time. ## Matches The page opens with **What this scan is looking for** — the search stated before the results it produced: the roles and seniority being fetched, where, the employer size floor, and the score a match must clear to be shown (your own Matches floor from Settings when you have set one, otherwise each search's threshold). Expand it to see every search with its title phrases grouped in plain words — *any of*, *and all of*, *but never* — the description filter, how many postings each search has found and when it last ran, plus the employers and titles that are never shown. It is read from your saved searches and profile, so it always describes what actually runs; a search that is paused says so rather than disappearing. Every posting your scans found or you saved with the browser extension. Click a **job title** to expand details: stated pay, key skills, what the role involves, what they're looking for, company industry and size, a link to the original posting, and — once you've tailored — your application package downloads. Click a **score** — here, on the Board, in a card's drawer, or on the Dashboard — to open the why-this-score panel: the written rationale, what the scorer found on seniority, location and pay, and links to the settings that produced it (interview answers, targeting, exclusions, a search's threshold, your Matches floor), with a re-score button for after you change one. A link to one job — from your agent, the [short list](/mcp/workspace/), or the search box — opens Matches with that job first, expanded and scrolled into view, marked *Opened from a link*, whatever the filters. Dismiss or pursue it and the next job takes its place. A job you already passed on shows **Restore to Matches** instead of Pursue and Dismiss. Prefer to review matches on your phone? [Jobably Swipe](/app/swipe/) deals the same postings as cards — right to keep, left to pass — and installs from the browser. Stated pay is captured wherever the posting declares it — salary ranges on scanned roles AND hourly or project rates on contract gigs you save with the extension. Jobably never estimates pay: if a posting doesn't state it, the field stays empty. When a **new match scores at or above your email alert level** (Settings → Notifications; default 8), Jobably emails you — as each scoring run lands, or **once a day** in one email if you prefer, or not at all. Never one email per job, and only the first time a posting is scored. This is a different number from a search's [score threshold](/scoring/threshold/), which decides what appears on Matches at all. Filters cover status, minimum score, source, engagement type (salaried vs contract work), **which of your searches found the match**, and **how recently it was found** (today, or within the last 3 / 7 / 14 / 30 days). Every match also has a **Re-score** button: watch it run live, and the fresh score lands on the card the moment it finishes. Finished statuses — including any errors — can be dismissed with the × so they don't linger. ## Pursuing Everything between "I want this" and "I sent it", in one place. Each row shows its **stage** — Queued → Drafted → Reviewed → Final → Applied — worked out from what actually exists (a studio session, a built document), and **one button** that changes with the stage: *Tailor resume* opens the studio, *Review & finalize* and *Decide & build* take you back to it, *Mark applied* stamps the date and moves the card to the Board. *Drop* returns the posting to Matches. Expand a row for the decision beside the evidence: **Why** (the score's rationale and what the posting is buying), the full **job description**, and the **documents**. ## Board A kanban view of your pipeline: Matched → Pursuing → Tailored → Applied → Screen → Interview → Offer, with a **Closed** list underneath. Drag cards between columns or use the keyboard-friendly "Move to…" select on each card — only legal next stages are offered. From Applied on, each card also has an **ended:** row (rejected / withdrawn / accepted); an ended application drops into Closed with its outcome and the stage it reached. The Pursuing and Tailored columns are mirrors of the [Pursuing](#pursuing) screen, where that work happens. Cards sitting in Applied for five days with nothing back show how long they have waited. Clicking a card opens a drawer with the full description, score rationale, downloads, your notes, and the complete status history. ## Resumes Every document, **grouped by application**: what Jobably wrote — the tailored resume, cover letter and prep packet, tagged *ours* — beside the final you actually sent, tagged *yours* (add it with **Upload** on the group). Each group says where the application stands. **Diff vs master** shows what the draft changed; **Diff vs your final** shows what you changed by hand. **Revise (keeps your decisions)** reopens the studio session; **Re-draft from scratch** discards those decisions and queues a fresh draft. Only a final you actually sent — the application is applied or beyond — can be marked an **exemplar**, which later drafts learn from. **Tailor for a job** opens the studio. Your master resume lives under **Profile → Master Resume** (upload new versions any time — old ones are kept and diffable; marking a version as an *exemplar* teaches future tailoring to imitate its style). The provenance guarantee applies everywhere: tailored documents only contain figures that exist verbatim in your master resume or accomplishment bank. Every tailored draft carries your **name, email and phone exactly as they appear in your master resume**, along with every employer, title and date. Those are facts about you, so nothing is asked to reproduce them. **Downloading a tailored document asks you to confirm you will read it first.** That is not a formality. The guards stop a draft inventing a figure; they cannot stop it being aimed at the wrong thing, or carrying a licence from another industry onto this application. See [what a tailored resume contains](/resumes/tailoring/). ### Gaps Beside every finished draft is the list of things the posting asked for that your record does not evidence. The draft never claims them. Each one is either something to add to your [accomplishment bank](/resumes/accomplishment-bank/), something for a cover letter, or a reason this role is not for you — and the third case is worth as much as the other two. ### Your accomplishment bank Tailoring can only use what it has been given, and a resume is one compressed snapshot of a career. The bank holds the rest, and you can **import your LinkedIn data export** into it from the Profile page — positions, dates and the description you wrote for each role. Nothing is replaced; entries are added and duplicates skipped. ## Searches Each search is a standing instruction: a boolean query, which sources to scan, a score threshold, and a cadence. The live preview shows exactly how your query compiles — including warnings about word-order traps. Searches pause rather than delete, so scan history is never lost. ## Profile **Profile** is the first item in the left navigation — a section that folds open to its pages: **Master Resume** (versions, diffs, exemplars), **Targeting** (the questionnaire — the single source of truth for scoring: engagement types, title keywords for and against, industries for and against), **Interview** (your answers, editable any time — they feed the rubric verbatim), **Exclusions**, **Accomplishments** (the bank), and [**Your Bio**](/resumes/bio/) — a few sentences in your own voice that your cover letters speak in. ## Billing, Settings These live in the **avatar menu** (top right): - **Billing** — plan status, trial countdown, usage, and the Stripe portal. - **Settings** — linked sign-ins, your own model keys (write-only) and the tailoring-model picker, search and email preferences (your own Matches score floor, the email alert level, how often match emails and the pipeline summary arrive — daily, weekly, or off), personal access tokens for the extension and AI agents — with a **connect card** that hands you the full ready-to-paste command for Claude Code, Codex, or Copilot — plus theme, data export, and account deletion. ## Getting help The **?** button in the header opens context-sensitive help for whichever screen you're on, including what each button does. The **speech-bubble** button (bottom right) sends us a bug report, idea, issue, or request from anywhere in the app. ## Managing searches https://docs.jobably.com/searches/managing/ > Cadence, thresholds, pausing, and why there is no delete. ## Score threshold Each search has a threshold (default 7). Postings below it are stored and scored but don't surface as matches — raise it when you're drowning, lower it when you're exploring. Nothing hidden is lost: the Matches page counts what's below the line and a **Show them** link reveals every one, badged "below threshold"; the Board's Matched column shows the same count. ## Pause, never delete Pausing stops the scans immediately, and Jobably remembers exactly where the search left off. Resuming picks up from that point — no gap, nothing fetched twice. There is deliberately no delete: deleting would throw away the search's memory and history. A paused search costs nothing and can sleep forever. ## Run now "Run now" queues an immediate scan across the search's sources. On a paused search it returns an error — resume first. ## Fetch filters Each search can also narrow what's *fetched* (which is what costs credits), all editable on the search form: - **Location** — defaults to United States; blank means worldwide. - **Years of experience** — 0-2, 2-5, 5-10 or 10+. This is the level filter company career sites use, and those are where most of your matches come from, so it is usually the one worth setting. - **Job-board levels** — Internship through Executive, as checkboxes. Job boards only; career sites ignore them, which is what years of experience is for. - **Work type** — full-time, part-time, contract, temporary, internship and so on. Defaults to full-time. - **Company size** — a minimum, a maximum, or both. - **Industries** — only these industries. - **Only these companies** — restrict a search to named employers. - **Skip these companies** and **skip titles containing** — per-search lists that ADD to the exclusions on your profile rather than replacing them. - **Skip staffing agencies** — on by default. - **Description must contain** — a boolean matched against the job description. It NARROWS the search: a posting must match your title query *and* this. Keep the title query broad if you use it. - **Remote only** — only roles that can be done remotely. Everything except the description filter is applied before you are charged, so a narrower search costs less rather than more. Titles on your profile-wide excluded keywords list are dropped at scan time too — that noise never reaches scoring. Jobs you save with the extension are never filtered. All of these are also available over the API and to a connected AI assistant, and behave identically in each. ## Sources - **ATS boards** (Greenhouse, Lever, Workday, …) — the highest hit-rate source in our proving run (27.8% strong-match rate vs 2.8% for general boards). On by default. - **Job boards** — broader but noisier. Also on by default. - **Extension** — anything you send from Chrome counts as its own source. ## Writing search queries https://docs.jobably.com/searches/query-syntax/ > How to write searches that find the right roles — and the traps Jobably catches for you. A search query is just words and a few simple connectors. Jobably turns it into a precise search behind the scenes and **always shows you what it understood**, both while you type and in scan health afterward. ## You usually don't have to write it Everywhere a query can be edited, Jobably shows a **guided builder** with three plain boxes: - **Match ANY of these** — one is enough (`"VP Engineering"`, `"Head of Engineering"`, …) - **Must ALSO contain** — every one required - **Must NOT contain** — excludes Add terms as chips; phrases are quoted and grouped for you, and the exact expression it builds is always shown underneath — what you see is what runs. An **Advanced** toggle switches to the raw expression for anything the boxes can't say. If a hand-written query is too intricate for the boxes, Jobably opens it in Advanced rather than ever silently changing what it means. The rest of this page explains the raw syntax — useful in Advanced mode or when reading what the builder produced. ## The basics | You write | What it means | |---|---| | `"VP Engineering"` | this exact phrase | | `A OR B` | either one | | `A, B, C` | any of them — a comma is a shorthand for OR | | `A AND B` (or just `A B`) | both | | `NOT recruiter` | leave these out | | `("VP Eng" OR "Head of Eng") NOT agency` | grouping with parentheses | Operators must be CAPITALISED. `OR`, `AND` and `NOT` combine terms; `or`, `and` and `not` in lower case are treated as ordinary words, so "Research and Development" searches for that department rather than for two separate things. As you type in the search form, a live preview shows exactly how your query was understood, plus any warnings. What you see in the preview is exactly what runs. ## Traps Jobably catches for you Job search queries have some surprising traps. These are real ones we hit — each is now handled automatically, with a warning so you know. ### Short words like "IT" Search engines ignore tiny common words ("it", "of", "the"). Without care, a search for `IT manager` would quietly become just `manager`. Jobably notices when a short word is written in ALL CAPS, treats it as an acronym, and keeps it — and tells you it did. ### Connectors inside quotes `"marketing OR sales"` searches for that literal phrase — the OR is just a word inside quotes. If you meant "marketing or sales," drop the quotes. ### NOT after a group `("A" OR "B") NOT C` excludes C from everything in the parentheses — the whole group, not just the last item. It works the way you'd read it. ### Phrases match word order exactly `"Director of Procurement"` will not match a posting titled "Procurement Director." When the order isn't essential, spell out both: `"Director of Procurement" OR "Procurement Director"`. Jobably warns you whenever a phrase is order-sensitive. ### A search that can't match anything If every word in your query is one the search engine ignores, the search would silently find nothing. Jobably flags this immediately instead of letting you wonder why results never arrive. ## How the rubric works https://docs.jobably.com/scoring/rubric/ > Your questionnaire becomes a deterministic scoring rubric with written rationales. Every posting is judged against a **rubric derived from your questionnaire** — not a keyword match, not a black box. The rubric is deterministic: the same questionnaire always produces the same rubric text, and you can read that text verbatim on your Profile page. ## What the rubric encodes - **Seniority floor** — "roles a full rung below score at most 4." - **Location as ranking, never filtering** — a stated location preference ranks results; it is explicitly forbidden from hiding them. - **Compensation floor** — stated-lower postings are marked; silence is unknown, not failure. - **Pay tiers** — instead of one floor, judge the **top** of the posted range in steps: for example, a range topping out below $200,000 scores at most 6, below $175,000 at most 4, and $200,000 or more is not capped. Set them under Profile → Targeting. - **Dealbreakers** — excluded title keywords reject regardless of employer. They match whole words and their plurals: "Commodity" also catches "Commodities". - **Engagement types** — salaried (W2) roles, contract/project work, or both. A contract gig is scored on its own terms (with your hourly-rate floor, if set), never penalized for not being salaried — and vice versa. Both are on unless you turn one off: only when you choose salaried-only (or contract-only) do postings of the other kind score at most 2, with the reason stated. - **Conditional rules** — a rule that applies only when its conditions hold. A condition tests the posting's title, description or location, and can require that it contains any of your words or none of them. Combine them: "individual-contributor titles count only in Charlotte" is a title that contains Manager or Analyst **and** a location that contains none of Charlotte, scored at most 5. Add rules under Profile → Targeting, or ask your agent to set them. - **Your own words come first** — the built-in location caps are a default. Where your conditional rules or your interview's location notes say otherwise, yours decide, and the rationale says which one applied. ## Scores come with reasoning Every score is accompanied by a written rationale ("Meets your seniority floor; platform scope matches; flag: hybrid 2 days"). A bare number you can't interrogate is worthless — the rationale is the product. Click any score in the app to read it, see the scorer's verdicts on seniority, location and pay, and jump to the setting that would change it. The same panel says what the posting is **buying** — a leader's seat, a doer's, or an analyst's — read from the responsibilities rather than the title. A Director title on a description that is all coordinating and processing is a doer posting, and the panel says so. This reading never moves the score; it tells tailoring which of your accomplishments to lead with for that posting. ## Ratings tune the rubric When you dismiss a posting you give a reason code **and a free-text note**. The note matters more than the code: "this is IC work dressed up as a VP title" teaches the rubric something "wrong_level" alone cannot. Resolved ratings are folded into rubric rebuilds and never re-flagged. ## Versioning Saving the questionnaire bumps the rubric version. Scores reference the version that produced them, so a rubric change never silently rewrites history. ## Tuning the threshold https://docs.jobably.com/scoring/threshold/ > When to raise or lower the match threshold, and where each number lives. Three numbers decide what you see and what you hear about. They answer different questions, so raising one never moves another. | Number | Where | What it decides | |---|---|---| | **Search threshold** (0–10, default 7) | each search's form | What that search's matches need to score to appear on Matches and the Board | | **Matches score floor** | Settings → Notifications | Your own floor for *every* search at once — set it and it outranks each search's threshold; leave it on the default and each search decides | | **Email alert level** (default 8) | Settings → Notifications | Which of the matches that appear also get emailed to you | Matches that score below the floor are never deleted — they're set aside. The Matches page tells you how many are hidden ("N matches scored below your search's score threshold") with a **Show them** link that reveals every one in place, badged so you can tell them apart. The Board's Matched column shows the same count and points back to Matches. **Raise it (8–9)** when the dashboard is noisy: you'll only see postings the scorer can defend strongly. Good during an employed, selective search. **Lower it (5–6)** when volume is low — early in a search, in a niche market, or when the scan health shows plenty fetched but few surfacing. The rationale text tells you *why* things score 5–6; if the reasons look wrong, fix the questionnaire rather than the threshold. **One floor for everything.** If you keep changing the same number on several searches, set the Matches score floor once in Settings instead. It applies to every search; the per-search thresholds stay as they were and take over again the moment you clear it. **Signals you're tuned wrong:** - Everything scores 8+ → your floor is set below your actual bar; raise the seniority floor or add dealbreakers. - Good roles score 4–5 with "below seniority floor" → your floor is set above what you actually want. - Zero results → usually the query, not the threshold — check how it was understood in [scan health](/getting-started/first-scan/#reading-scan-health). ## Your accomplishment bank https://docs.jobably.com/resumes/accomplishment-bank/ > Why a wider record produces a better tailored resume, and how to import your LinkedIn history into it. Tailoring can only work with what you have given it. Your master resume is one compressed snapshot of your career — a page or two, edited down over years to fit. Everything that did not fit is still true, and still the thing that might win you a particular role. A tailored draft cannot reach for a result it has never been shown. The **accomplishment bank** is where the rest of your record lives. It feeds tailoring alongside your master resume, and it is subject to the same [provenance rule](/resumes/provenance/): a figure may only appear in a tailored draft if it appears here or in your master, verbatim. ## Why this changes the output A posting asks for capital-equipment procurement. Your standing resume leads with software and services, because that is what your last three roles were about — so a tailored draft leads there too, and reads slightly beside the point. If the bank holds the equipment-financing result from six years ago, the draft can lead with that instead. Same career, same truth, better aimed. The difference is not the wording; it is whether the fact was in the room. This is also why the [gap list](/resumes/tailoring/) beside a finished draft is worth reading. Each gap is either something to add to the bank, or an honest reason this role is not for you. ## Import from LinkedIn Your LinkedIn profile usually holds far more than your resume: the description you wrote for each role, at the time, before it was edited down. 1. In LinkedIn, go to **Settings → Data privacy → Get a copy of your data**. 2. Choose the **full archive** rather than a specific file. LinkedIn splits exports, and only the full archive contains your positions. 3. LinkedIn emails you a zip, usually within a few minutes to a day. 4. In Jobably, go to **Profile → Accomplishment bank → Import from LinkedIn** and upload the zip, unmodified. Jobably reads the role descriptions and adds them as accomplishments. **Nothing is replaced.** Entries are added to what you already have, duplicates are skipped, and you are told how many came in, how many roles had no description to learn from, and whether anything was dropped at the 1,000-entry limit. ### What it does and does not read It reads your **positions** — company, title, dates, and the description you wrote. It also reads your headline, if present. It does not read your connections, messages, or anything else in the archive. Skills and endorsements are deliberately skipped: a skill name on its own is not an accomplishment, and adding bare terms to the bank would weaken exactly the guard that keeps tailored drafts honest. ### If you would rather not export The bank is a plain list. You can paste into it directly on the same page — one accomplishment per line, with the real figures. Old resumes, performance reviews and project write-ups are all good sources. ## Writing good entries - **Keep the numbers exact.** `$4.8M`, `20%`, `1,300+ suppliers`. Figures are what tailoring can never invent, so an entry without them is weaker. - **One accomplishment per line.** A paragraph covering three results is harder to select from than three lines. - **Include the ones you have moved on from.** The point of the bank is the work that no longer fits your resume but still fits some job. ## Your Bio https://docs.jobably.com/resumes/bio/ > A few sentences in your own voice — the voice of your cover letters, and a draft for your LinkedIn About. **Profile → Your Bio** holds one short self-description in your own words: a headline and three to six sentences. Two things use it. - **Cover letters speak in it.** When a draft comes with a cover letter, the letter is written in the voice of your bio and draws on it where it fits the role. - **It is a draft for your LinkedIn About.** Copy it out when you are ready; nothing in Jobably ever posts anywhere. ## Evidence, not adjectives A bio that could be anyone's is not worth having. The test: swap the name, and does it still read true? If so, it says nothing. Lead with what only you could write — the size of the portfolio you ran, the team, the years at a level, the savings, the named programs and credentials. **Propose a draft from my record** writes one for you from your master resume, your accomplishment bank and your interview answers, in the first person. It is a proposal: it appears above the editor and nothing is saved until you click *Use this draft*, after which it is yours to edit. ## The same no-fabrication rule as a resume line Saving runs the check every tailored line passes: a year, a figure or a number that does not appear in your master resume or accomplishment bank is refused, and the refusal names it — *"45" does not appear in your record*. That is on purpose. The bio feeds documents that carry your name, so a fact has to be in your record before it can be in your bio. If it is true and missing, add it to the [accomplishment bank](/resumes/accomplishment-bank/) first. Prose is free — the rule is about facts, not phrasing. ## Provenance rules https://docs.jobably.com/resumes/provenance/ > Why a tailored resume can never invent a figure. Jobably's tailoring runs under one hard rule: > **Every figure in a tailored resume — every dollar amount, percentage, > headcount, and large number — must exist verbatim in your master resume or > accomplishment bank.** Not paraphrased. Not recalculated. Verbatim. ## How it's enforced After generation, the draft is checked mechanically. Any figure that cannot be traced to your documents rejects the draft — it is regenerated, and you never see the fabricated version. This is a code-level guard, not a prompt instruction the model might ignore. ## Counts are figures too A number of things — direct reports, teams, deals, suppliers, years — has no dollar sign or percent, but it is the first claim an interviewer tests. A count is used exactly as your record states it: the same number, with the same thing it counts ("5 direct reports" never becomes "4 direct", and "16 teams" never becomes "16 suppliers"). Years in a date, and ranks like "Tier 1", are not counts. When a number changes, update your record: the tailor reads the bank and master you last saved, so a line you remove or correct never comes back, even if an older resume you sent still says it. ## Gaps are surfaced, not papered over If a posting wants something your history doesn't evidence, the tailored output lists it as a **gap** instead of writing around it. Gaps are useful — you can address them in a cover note or prepare for them in the interview. An invented claim, discovered under one pointed follow-up question, ends the interview. ## Feed the bank The accomplishment bank (Profile page, one accomplishment per line, real figures included) is the guard's whitelist alongside the master resume. The richer it is, the more the tailor can honestly say. ## Reviewing a draft before you send it https://docs.jobably.com/resumes/reviewing/ > Approve, decline, or reword each suggested change — the draft only ever changes where you said yes. A finished draft is a proposal, not a verdict. Beside every tailored resume, Jobably now shows you what the posting asks for that the draft doesn't say — sorted by whether your own record can honestly support it — and lets you act on each item in place, without starting over. ## The three groups - **In your record.** The posting wants a phrase, your record genuinely carries the substance, and the draft just doesn't say it yet. Each item shows the evidence — a quote from your own material — and a suggested wording. These start selected. - **Judgement calls.** Your record comes close, but whether the claim is fair is your call, not ours. The evidence is shown; these start unselected. Select the ones you stand behind, and reword them if the suggestion overreaches. When there is a real limit on how far a phrase may go, a **caution** names what must not be claimed — and the revision is held to it, not just you. - **Not in your record.** The posting wants something your record simply doesn't contain. There is no checkbox here on purpose: Jobably never adds a claim it cannot source. This list is still useful — it's what to go gather evidence for, or to address in the interview instead. ## Applying your decisions Choose your items, optionally type your own wording for any of them, and apply. Only what you approved changes, and only where it belongs — every line you didn't touch comes back exactly as it was. The result is a new version of the draft (your earlier version is kept, as always). Two honest outcomes to know about: - **Declined changes.** If an approved change can't be made truthfully — the wording would claim more than your record states — it is declined and you're told why. A declined change is the system working, not failing. - **Reverted edits.** If the rewrite touched anything you didn't ask it to, those lines are put back automatically and the draft tells you it happened. The [provenance rules](/resumes/provenance/) hold in full during a revision: figures, dates, employers and titles are never altered, and nothing is added that your record can't back. ## What else the draft tells you Beside every draft you'll also see, when they apply: - **Sections kept in your own words** — where tailoring left your text untouched, it says so and says why for each one: there was nothing to tailor (a contact line, a one-line banner), the time budget ran out before that section's turn, or every rewrite failed the no-fabrication guards. A section skipped on purpose and one that ran out of time are different facts, and they are never shown as the same one. - **Style notes** — advisory observations (a repeated opening verb, a phrase hiring managers see everywhere). Notes, not corrections: nothing was changed or blocked because of them. - **Orientation** — what the posting reads as (a leader's seat, a doer's, an analyst's) and what the draft reads as by the same measure, with a flag when they differ. Tailoring orders the bullets within each role to lead with the posting's orientation — order and emphasis only; the same bullet sits in a different place on different applications, and no bullet changes what it claims to fit. A mismatch is something to consider, not a defect: leading with scope on a doer posting may be exactly your intent. - **The scorecard** — a mechanical summary of how the draft measures up (contact details present, every employer accounted for, the posting's language echoed), so you can compare versions on more than a feeling. ## The tailoring studio https://docs.jobably.com/resumes/studio/ > Tailor your resume for one posting with Jobably beside you — line by line, role by role, every change yours to accept, reword, or keep. The studio is where you tailor a resume **with** Jobably rather than waiting for one to be tailored for you. Open it from **Resumes → Tailor for a job** in the left navigation, or from **Tailor resume** on any match or board card. It walks your resume line by line — and, inside your experience section, role by role — and proposes changes with the reason each one serves this posting. Nothing changes until you say so. ## Three steps ### 1. Review Pick the master version to start from. If you have [pinned one for the Studio](/resumes/versions/), the studio starts there; otherwise it starts from your latest upload. Then **Review my resume**. Jobably reads the posting first — what it repeats, what it requires, the level it is written for, and where it says the same thing as your resume in different words. Then it reviews each section. Sections appear as they finish, so you can start deciding while the rest is still running; leaving the page does not stop it. On your own connected model ([BYOK](/byok/)), the review also brings **structure and layout recommendations** — which section should lead for this posting, what to tighten, how to re-aim the summary, which of your existing competencies to put first — and a headline aimed at the role. On the included model these are skipped, and the studio says so. ### 2. Decide Every line of your resume is shown as it stands. Where Jobably has a proposal, it sits beneath the line with the reason and the posting phrases it works in, and you choose: - **Use this** — the proposed wording replaces yours in this draft. - **Keep mine** — your line stands. - **Edit** — write your own version. Any line, proposed or not, can be edited or left out of this draft. Your master resume is never changed; every decision applies to this posting's draft only. **Length is decided before the review, and you can overrule it.** The same budget a one-shot draft runs under applies when a session opens: up to 12 bullets for your current role, 6 for roles that ended in the last ten years, fewer for older ones, or the limit you set under Settings → Tailoring. Each bullet left out is marked *left out for length* in its section — put it back if it matters for this posting, and the built document reports what was left out. **A proposal keeps your words.** A proposal that swaps in the posting's word for one your own line never uses (the posting says *supplier*, your line says *vendor*), or that appends the posting's own sentence to one of yours, is held back with the reason — in every section, the summary and your skills sections included. You can still write the word yourself by editing the line — the rule is for the model, not for you. **A proposal keeps your opener.** A rewording that softens how a line opens (*Drove* becoming *Contributed to*) is held back; the line's own opener stays. **Competency rows are reordered, never rewritten.** A proposal for a row of competencies may only reorder the terms you listed, leading with what the posting asks for. One that adds, renames or drops a term is held back. **A line appears once.** A bank entry already in one section of this draft is not proposed for another, and the build refuses a copy that slips through. The summary admits one added paragraph at most — the team-size line when the posting asks for one — and when every paragraph of your summary opens with a label, an addition is asked to open with one too. **What the posting opens with leads your current role.** When the posting's first sentence about the role names a duty and a bullet in your current role answers it, the built document places that bullet in the role's top three and reports the move. **Team scope, when the posting asks for it.** When the posting asks for leading people and your record states your team size, the summary review is asked to state it, with the figures exactly as your record has them, and proposes the line from your bank if it did not. Additions come from your [accomplishment bank](/resumes/accomplishment-bank/): when the posting asks for something this section could evidence, the relevant bank entry is proposed as a new line, quoted beside it, for you to add, skip, or reword. Two things the studio will not do: - **Apply a proposal you have not decided on.** An undecided proposal is not applied; your own words are always the default. - **Show a proposal as if it were safe when it is not.** Every proposal is checked against your record before you see it — every figure verbatim, every number on a line unchanged, every date as written, every sentence traceable to your resume or your bank. A proposal that fails is shown as **held back**, with the reason, and cannot be selected. The same checks run on anything you type: a number your record does not carry is refused where you typed it, with the reason, not at the download. ### 3. Build **Build the document** composes the draft from your decisions, checks every changed line once more, and renders it into your own Word file — your fonts, margins and bullets — with a matching cover letter if you leave that box ticked. The result appears on the job's card and under Resumes like any tailored resume, moves the job to **Tailored** on your board, and downloads through the same [read-before-you-send step](/resumes/reviewing/) as every tailored file. Beside the built document you get what the posting still asks for that your record does not evidence, and the [review panel](/resumes/reviewing/) for one more approve-and-decline round on the finished draft. Change any decision and build again for a new version; earlier versions are kept. ## From a connected assistant The same session is reachable over [MCP](/mcp/tools/): `open_tailoring_session`, `review_tailoring_session`, `get_tailoring_session` (every proposal with its reason and guard verdict), `decide_tailoring`, `build_tailored_resume`, and `discard_tailoring_session`. It is one session per job, so you can start in Claude and finish on the page, or the other way round. `tailor_resume` still queues the one-shot background draft. ## Starting over **Start over** discards the proposals and decisions in a session and begins from the master again. Documents you already built stay on the job. Switching to a different master version also starts a fresh session, because your decisions were made on the lines of the version you were looking at. ## What a tailored resume contains https://docs.jobably.com/resumes/tailoring/ > How Jobably rewrites your resume for one posting, what it never changes, and what it tells you it could not cover. Tailoring starts with **your** resume: your sections stay under your headings and in their original order unless you reorder them in the studio. For supported Word masters, Jobably edits text inside the original document, keeping its fonts, alignment, tables, lists, and headers and footers. Longer wording can still change line breaks or page count. If an edit cannot safely retain the original layout, the download uses a standard layout and the app tells you. PDF and plain-text masters also use that standard Word layout. The place to do it is the [tailoring studio](/resumes/studio/), where every change is proposed to you line by line and applied only when you say so. Everything below describes what a tailored draft contains and never changes, whichever way it was produced. ## What stays exactly as you wrote it - **Your name, email and phone.** Taken straight from your master resume and preserved in the draft. These are facts about you, so nothing is ever asked to reproduce them. - **Your section headings, and their order.** If your resume runs Summary, Experience, Projects, Education, so does the draft unless you reorder it yourself in the studio. - **Every figure.** Dollar amounts, percentages, headcounts and dates appear exactly as they do in your record, or they do not appear at all. See [provenance rules](/resumes/provenance/). - **Employers, titles and dates.** Tailoring reorders and rewords what sits under them; it never changes the record itself. ## What changes - **Emphasis.** The results most relevant to this posting move up. - **Wording.** Bullets are rewritten toward the posting's own language — the same accomplishment, said the way this employer talks about it. But a word the posting uses and your record never does may not replace yours: that line keeps your own wording, and the notes beside the draft name the word that was turned away. The one exception is the posting spelling out an acronym your record uses ("vendor risk management" for "TPRM"). - **Length.** Experience bullets are kept by role: up to 12 for your current role, 6 for roles that ended in the last 10 years, 3 within 15, and one for anything older — never none. The bullets closest to the posting stay, and a weak line with no figure goes first. Add an overall limit, or keep every bullet, under **Settings → Tailored resume length**. Each bullet left out is listed beside the draft, so you can put one back by hand. - **Opening verbs.** Two bullets side by side never start with the same verb. Jobably reorders them to keep repeats apart, never rewords them. - **Your headline.** Angled at the role, while staying true. - **Core competencies.** Reordered for this posting, and filtered: a competency appears only if it is already in your master resume. A skill your record cannot back is a claim you would have to defend cold in an interview, so it is dropped rather than included. Sections with nothing to tailor — a contact line, an availability note — are passed through untouched. ## The gap list Beside a finished draft you get **what this posting asks for that your record does not evidence** — and, for items your record CAN support, a [review step](/resumes/reviewing/) that lets you approve, decline, or reword each suggested addition before it touches the draft. This is deliberately not hidden. A tailored resume that quietly papers over a missing requirement is the one that gets found out in the interview. Each gap is one of three things: 1. Something you **do** have that is simply not written down — add it to your [accomplishment bank](/resumes/accomplishment-bank/) and tailor again. 2. Something you could address in a cover letter or conversation. 3. A real mismatch, and a reason not to spend an afternoon on this application. The third case is a feature. Knowing a role is not for you before you apply is worth more than a well-worded draft for it. ## If a draft comes back partly untailored Long resumes are tailored section by section. If a section could not be rewritten within the time budget, the draft ships **your original wording** for that section rather than something rushed or invented, and says which sections those were — with the reason beside each name, so a section that had nothing to tailor is never mistaken for one that ran out of time. If **no** section could be tailored, there is no draft to show: the run fails with a "try again" rather than handing you your own resume back as if it were tailored. The usual cause is running on the included model. Connecting your own API key under [BYOK](/byok/) is markedly faster and degrades far fewer sections. ## Versions & diffs https://docs.jobably.com/resumes/versions/ > Master resume versioning, diffs, exemplars, and downloads. ## Master versions Every upload creates a new version; nothing is overwritten. **Profile → Master Resume** lists all versions with downloads and a **diff** against the previous version, so you can see exactly what changed between v1 and v2. (Tailored drafts live under **Resumes**, grouped by application, beside any final of your own you upload there — and a final you actually sent can be marked an exemplar that later drafts learn from.) ## Pinning a version for the Studio **Pin for Studio** makes the [tailoring studio](/resumes/studio/) start from that version instead of your newest upload. While a newer master exists, the page says so, because the Studio keeps reading the pinned one. Unpin it to go back to the newest. ## Exemplars An exemplar tells the tailor "write like this one": a final you actually sent, marked under **Resumes**. Exemplars shape tone and structure; they never bypass the [provenance rules](/resumes/provenance/). Pinning a master version does not make it an exemplar. ## Tailored artifacts Tailored resumes are generated per posting as `.docx`, listed on the Resumes tab and on each job's card, downloadable any time — including while your account is read-only. ## Board & workflow https://docs.jobably.com/board/workflow/ > The pipeline states, drag-and-drop, notes, history, and how applications age. The board is a kanban over a strict status machine: ``` matched → pursuing → tailored → applied → screen → interview → offer ``` plus **dismissed**, which is reversible (dismissed → matched) — reconsidering is normal. **Screen** is a recruiter screen or phone call; it is optional — an application can go straight to Interview. Illegal jumps are rejected by the server (e.g. matched → interview), and `date_applied` is stamped exactly once, the first time a posting reaches applied. History is append-only — no one, including Jobably, can rewrite it. ## Two axes: the stage reached, and what came back An application does not only move forward. From Applied on, every card carries an **ended:** row — *rejected*, *withdrawn*, *accepted* — because their decision is a different fact from how far you got. A card that ends leaves the columns for the **Closed** list under the board, which says both: *rejected · reached interview*. "Rejected at applied" (no human read it) and "rejected after an interview" are not the same outcome, and the list never blurs them. Closed stays closed; the history holds the trail. The Dashboard's **What came back** tiles are the tally: still open, rejected, withdrawn, accepted, the average days to hear back, and the share of applications that got an answer at all. ## Moving cards Drag cards between columns, or use the **Move to…** menu on each card — it lists only the legal next stages and is fully keyboard-accessible. ## Card drawer Click a card for the full picture: the job description, score + rationale, tailored artifacts, notes, and the complete status history. ## Pursuing and Tailored are mirrors The work between Matched and Applied — tailoring, reviewing, building the document, marking it applied — happens on the **Pursuing** screen, where each posting shows its stage and one action. The Board's Pursuing and Tailored columns show the same cards and point across. ## Waiting on a reply Five days after a posting reaches **applied** with nothing back, its card shows how long it has waited and it appears in the dashboard's "Waiting on a reply" list. This is an age, not a prompt to chase: on most application portals there is nothing to send, so the count is for knowing when to write one off. It clears the moment the status changes. ## Chrome extension https://docs.jobably.com/extension/install/ > Install, connect with a token, and send any posting to Jobably. The extension covers what the scanners can't: any job page you open in Chrome. It reads a page **only when you click it** — no background access, no other tabs, no analytics. ## Install Install from the [Chrome Web Store](https://chromewebstore.google.com/detail/ehefnaajjdncohccmlgemcplonmjoffj) — one click, updates arrive automatically. (For development builds only: download the extension bundle, open `chrome://extensions`, enable **Developer mode**, and **Load unpacked** on the `dist` folder.) ## Connect your account 1. In the app: **Settings → Integrations → Personal access tokens → Create token** with the **ingest** scope. The full token is shown exactly once. 2. Click the Jobably icon in Chrome and paste the token. The token is stored only in your browser. It can ingest postings — it cannot read your pipeline, billing, or keys — and you can revoke it any time from Settings. ## Use it On any job posting, click the icon: - **Save to Jobably** — parses the posting (structured JSON-LD when the site provides it, text extraction otherwise), dedupes it against your matches, and scores it against your rubric. Stated pay comes along when the page declares it — including hourly and project rates on contract gigs. - **Save + tailor** — also queues a tailored resume. - **Save + prep** — also queues an interview prep packet. ## Permissions, explained | Permission | Why | |---|---| | `activeTab`, `scripting` | Read the posting on the page you invoked it on — nothing else, never in the background | | `storage` | Keep your token in this browser | | `api.jobably.com` | Deliver the posting to your account | ## Bring your own agent https://docs.jobably.com/mcp/connect/ > Connect ChatGPT, Claude, Codex, Copilot, Cursor or any MCP client to your Jobably account. Jobably works with the AI agent you already use. Connect it once and it can read your daily short list, search for jobs, tailor your resume, show you the draft, record what you applied to, and build you reports — all inside your account, all on the record. There are two ways to connect, depending on the agent: - **Chat apps** (ChatGPT, claude.ai, Claude Desktop) — add Jobably as a connector and sign in. You approve the connection in Jobably. Nothing to copy. - **Coding and command-line agents** (Claude Code, Codex, GitHub Copilot, Cursor) — paste one command with a personal token. The address is the same for every agent: `https://mcp.jobably.com/mcp` ## Chat apps: sign in, approve, done ### ChatGPT 1. In ChatGPT, open **Settings → Apps & Connectors → Advanced** and turn on developer mode (your workspace admin may need to allow it). 2. Back in **Apps & Connectors**, choose **Create** and enter the address `https://mcp.jobably.com/mcp`. Name it Jobably and choose OAuth. 3. ChatGPT sends you to Jobably. Sign in if asked, check what it may do, and choose **Allow**. ### claude.ai and Claude Desktop 1. Open **Settings → Connectors → Add custom connector**. 2. Enter the address `https://mcp.jobably.com/mcp` and choose **Add**. 3. Choose **Connect**. Claude sends you to Jobably; sign in if asked and choose **Allow**. ### Any other agent that supports sign-in Give it the address. If it supports OAuth for MCP servers it finds the sign-in on its own and sends you to Jobably to approve it. ### What you approve The approval screen shows the agent's name, where it will send you back to, and what it may do: - **Read your job search** — matches and scores, your short list, searches, the board, resumes and drafts, reports, and the activity record. - **Act on it for you** — move jobs on the board, record applications, run searches, tailor resumes (this uses your plan's credits), and build lists and reports. Untick the second to connect an agent that can only look. You can disconnect any agent at any time in **Settings → Integrations → Connected agents**. ## Coding and command-line agents: one command 1. In the app, open **Settings → Integrations → Personal access tokens** and create a token. Give it `read`, and `write` if the agent should act. 2. The **Connect your AI assistant** card right below shows the full command for your agent and operating system with the new token already filled in. Copy, paste, done. The same commands, for reference: ### Claude Code ```bash claude mcp add jobably --transport http https://mcp.jobably.com/mcp \ --header "Authorization: Bearer YOUR_TOKEN" ``` Claude Code also supports sign-in: leave out the `--header` line and run `/mcp` inside Claude Code to connect through the approval screen instead. ### Codex ```bash codex mcp add jobably -- npx -y mcp-remote https://mcp.jobably.com/mcp \ --header "Authorization: Bearer YOUR_TOKEN" ``` ### GitHub Copilot (VS Code) ```bash code --add-mcp '{"name":"jobably","type":"http","url":"https://mcp.jobably.com/mcp","headers":{"Authorization":"Bearer YOUR_TOKEN"}}' ``` ### Cursor and other clients (JSON config) ```json { "mcpServers": { "jobably": { "type": "http", "url": "https://mcp.jobably.com/mcp", "headers": { "Authorization": "Bearer YOUR_TOKEN" } } } } ``` ## New to Jobably? Let your agent set you up You don't need to set anything up in the app first. Connect your agent — signing in creates your Jobably account if you don't have one — then ask: > "Set me up on Jobably. Here's my resume." Attach or paste your resume. Your agent uploads it, and Jobably proposes a profile and a first search from it: your current titles, the level they sit at and where your current role is based, each with where it was read. Your agent shows you the proposal and asks what a resume can't say — the lowest salary worth seeing, remote, hybrid or on-site, and any industries to favour or avoid. Say yes and it saves everything in one step. The first scan runs immediately; your first matches arrive within minutes, scored against your profile. Ask "How's my Jobably setup going?" at any point and it tells you the next step. Nothing is saved until you approve it, and everything stays editable in the app afterwards. The same quick start is in the app's [onboarding](/getting-started/onboarding/#quick-start-one-click-from-your-resume). ## A good first thing to ask > "What's on my Jobably short list today? Tailor a resume for the best one > and show me the draft." Then, once you have applied: > "I applied to the Acme role this morning — mark it applied." See [Your agent's workspace](/mcp/workspace/) for the daily short list, lists, reports and the activity record. ## Spending and changing things Every tool says up front whether it is **free** or **metered**. Anything that spends your plan's credits or changes your account needs the agent to pass `confirm: true` — a call without it is refused with an explanation, so a well-behaved agent asks you first. Chat apps also see which tools only read, so they can ask before the others. Your agent never applies to a job for you. You apply; your agent records it. ## Passing arguments correctly Arguments are a plain JSON object. Two calls, shaped exactly right, cover the mistakes that actually happen: ```json { "name": "add_accomplishments", "arguments": { "bank": [ "Negotiated $4.8M savings, 20% on $24M of capital equipment spend.", "Cut sourcing cycle time 38% across a $40M program." ] } } ``` `bank` is a **real JSON array** — not a quoted string containing one — with **one accomplishment per element**. Commas stay INSIDE an entry: the first element above is one accomplishment, not three. Never pre-split text on commas; split on line breaks. ```json { "name": "create_search", "arguments": { "name": "VP Engineering — remote, US", "query": "\"VP Engineering\" OR \"VP of Engineering\"", "location": "United States", "experienceLevel": "5-10,10+", "excludeTitles": ["recruiter"], "remoteOnly": true, "cadenceHours": 24, "confirm": true } } ``` `confirm` is the **boolean `true`**, not the string `"true"`. Numbers may arrive as strings and coerce (`"24"` means 24), but unparseable values error loudly rather than being replaced with a default. Ids (`jobId`, `searchId`, `listId`, `reportId`, `operationId`, `artifactId`, `resumeId`) always come from a prior call — never invent one. Every tool in the [reference](/mcp/tools/) carries an example call validated against the live schema in CI — copy its shape. ## What an agent cannot do Signing in, creating tokens, approving other agents, model keys, billing and deleting the account happen only in the app, signed in as you. An agent can never create credentials or spend money outside the metered tools. ## Searching for someone else `search_jobs` runs one-off searches that are **never scored against your profile** and never saved. With `forSomeoneElse: true`, level and size filters are relaxed entirely — useful when a friend asks "can you look for staff-level roles for me?" without your rubric contaminating their results. ## Tool reference https://docs.jobably.com/mcp/tools/ > Every MCP tool, its cost class, every argument it accepts, and a correct example call. Generated from the running tool registry, so this page always matches what the server actually accepts. ## Passing arguments correctly Every rule below exists because an agent got it wrong in the field. The per-tool examples further down are validated against the live schemas in CI — copy their shapes. - **Arrays are real JSON arrays**, never a quoted string: `"bank": ["entry one", "entry two"]` — not `"bank": "[\"entry one\"]"`. - **One accomplishment per array element.** Commas are punctuation INSIDE an entry — "Negotiated $4.8M savings, 20% on $24M of spend" is ONE entry. Never pre-split text on commas. - **`confirm` is the boolean `true`**, not the string `"true"`. Tools that spend quota or change state refuse without it and tell you so. - **Numbers may arrive as strings** (`"cadenceHours": "24"` coerces to 24), but garbage errors loudly rather than being replaced with a default — prefer real numbers. - **Ids come from prior calls** — `jobId` from `list_matches`, `searchId` from `list_searches`, `operationId` from the tool that queued the work. Never invent one. - **`update_search` merges**: send only what changes; every filter you do not mention is preserved. | Tool | Cost | Confirm? | |---|---|---| | [`search_jobs`](#searchjobs) | metered: one source query per pool | — | | [`count_jobs`](#countjobs) | free — costs NO job credits | — | | [`list_matches`](#listmatches) | free | — | | [`get_job`](#getjob) | free | — | | [`set_job_status`](#setjobstatus) | free, state-changing — requires confirm: true | yes | | [`rate_job`](#ratejob) | free | — | | [`tailor_resume`](#tailorresume) | metered: tailor quota — requires confirm: true | yes | | [`revise_tailored`](#revisetailored) | metered: tailor quota — requires confirm: true | yes | | [`list_tailoring_sessions`](#listtailoringsessions) | free | — | | [`open_tailoring_session`](#opentailoringsession) | free | — | | [`get_tailoring_session`](#gettailoringsession) | free | — | | [`review_tailoring_session`](#reviewtailoringsession) | metered: tailor quota — requires confirm: true | yes | | [`decide_tailoring`](#decidetailoring) | free | — | | [`build_tailored_resume`](#buildtailoredresume) | metered: tailor quota — requires confirm: true | yes | | [`discard_tailoring_session`](#discardtailoringsession) | free, state-changing — requires confirm: true | yes | | [`get_prep_packet`](#getpreppacket) | metered: prep quota — requires confirm: true | yes | | [`get_board`](#getboard) | free | — | | [`move_card`](#movecard) | free, state-changing — requires confirm: true | yes | | [`list_searches`](#listsearches) | free | — | | [`create_search`](#createsearch) | metered: recurring scans every cadence — requires confirm: true | yes | | [`pause_search`](#pausesearch) | free, state-changing — requires confirm: true | yes | | [`get_operation`](#getoperation) | free | — | | [`list_accomplishments`](#listaccomplishments) | free | — | | [`add_accomplishments`](#addaccomplishments) | free | — | | [`update_search`](#updatesearch) | free, state-changing — requires confirm: true | yes | | [`run_search_now`](#runsearchnow) | metered: one source query per pool — requires confirm: true | yes | | [`get_search_health`](#getsearchhealth) | free | — | | [`rescore_job`](#rescorejob) | metered: score quota — requires confirm: true | yes | | [`add_job_note`](#addjobnote) | free | — | | [`declare_applied`](#declareapplied) | free, state-changing — requires confirm: true | yes | | [`get_exclusions`](#getexclusions) | free | — | | [`set_exclusions`](#setexclusions) | free, state-changing — requires confirm: true | yes | | [`list_resumes`](#listresumes) | free | — | | [`list_operations`](#listoperations) | free | — | | [`get_tailored_document`](#gettailoreddocument) | free | — | | [`get_master_resume`](#getmasterresume) | free | — | | [`mark_exemplar`](#markexemplar) | free, state-changing — requires confirm: true | yes | | [`delete_tailored_document`](#deletetailoreddocument) | free, state-changing — requires confirm: true | yes | | [`get_usage`](#getusage) | free | — | | [`get_shortlist`](#getshortlist) | free | — | | [`regenerate_shortlist`](#regenerateshortlist) | free, state-changing — requires confirm: true | yes | | [`list_applications`](#listapplications) | free | — | | [`show_tailored_resume`](#showtailoredresume) | free | — | | [`query_jobs`](#queryjobs) | free | — | | [`get_insights`](#getinsights) | free | — | | [`get_activity`](#getactivity) | free | — | | [`list_lists`](#listlists) | free | — | | [`create_list`](#createlist) | free, state-changing | — | | [`get_list`](#getlist) | free | — | | [`update_list`](#updatelist) | free, state-changing | — | | [`delete_list`](#deletelist) | free, state-changing — requires confirm: true | yes | | [`add_to_list`](#addtolist) | free, state-changing | — | | [`remove_from_list`](#removefromlist) | free, state-changing | — | | [`list_reports`](#listreports) | free | — | | [`get_report`](#getreport) | free | — | | [`save_report`](#savereport) | free, state-changing | — | | [`delete_report`](#deletereport) | free, state-changing — requires confirm: true | yes | | [`get_profile`](#getprofile) | free | — | | [`set_questionnaire`](#setquestionnaire) | free, state-changing — requires confirm: true | yes | | [`set_preferences`](#setpreferences) | free, state-changing — requires confirm: true | yes | | [`upload_master_resume`](#uploadmasterresume) | free, state-changing — requires confirm: true | yes | | [`get_started`](#getstarted) | free | — | | [`apply_setup`](#applysetup) | metered: recurring scans every cadence — requires confirm: true | yes | | [`score_new_matches`](#scorenewmatches) | metered: scores every unscored match against the rubric — requires confirm: true | yes | ## search_jobs Run a ONE-OFF job search. Results are raw postings and are NEVER scored against the owner's profile. The owner's exclusion lists and stated country are applied to the results unless forSomeoneElse is true. Set forSomeoneElse: true when searching on behalf of somebody else — it turns the owner's standing filters off. Nothing is saved. Prefer create_search when the user wants this run repeatedly. The response reports `truncated` and (when it can) `totalAvailable` — never present a truncated list as the whole market. A one-off is capped at 100 per pool and has no memory, so on a high-volume query it is a SAMPLE; create_search is what keeps up with one over time. count_jobs sizes a query for free. | Argument | Type | Notes | |---|---|---| | `query` | string | Boolean title search, e.g. "VP Engineering" OR "Head of Platform". A comma separates alternatives (reads as OR). | | `location` | string | e.g. "United States", "Charlotte, NC". Omit to search worldwide, which costs more and returns roles the candidate cannot take — set it unless they truly want worldwide results. | | `seniority` | array of string | LinkedIn vocabulary: Internship, Entry level, Associate, Mid-Senior level, Director, Executive. Applies to the boards pool only — the ATS pool ignores it, so use experienceLevel there. · one of: `Internship`, `Entry level`, `Associate`, `Mid-Senior level`, `Director`, `Executive` | | `experienceLevel` | string | Years of experience: '0-2', '2-5', '5-10', '10+' — one, or several as an array or comma string ('5-10,10+'). This is the ONLY level filter the ATS pool honours; without it an ATS search returns junior roles alongside senior ones. An unrecognised value is not applied and is reported in `warnings`. · one of: `0-2`, `2-5`, `5-10`, `10+`, `5-10,10+`, `2-5,5-10`, `0-2,2-5` | | `employmentType` | array of string | FULL_TIME, PART_TIME, CONTRACTOR, TEMPORARY, INTERN, VOLUNTEER, PER_DIEM, OTHER. Defaults to FULL_TIME — set it explicitly to reach contract roles. · one of: `FULL_TIME`, `PART_TIME`, `CONTRACTOR`, `TEMPORARY`, `INTERN`, `VOLUNTEER`, `PER_DIEM`, `OTHER` | | `minEmployees` | number | Minimum organisation headcount. · range 1– | | `maxEmployees` | number | Maximum organisation headcount. · range 1– | | `industry` | array of string | Organisation industries to include. | | `companies` | array of string | Only these organisations, by name. The inverse of excludeCompanies — filtered at the supplier, so it costs less rather than more. | | `excludeCompanies` | array of string | Drop these organisations from the results. Stacks ON TOP of the owner's standing exclusion lists; it never replaces them. | | `excludeTitles` | array of string | Drop postings whose title contains any of these (case-insensitive substrings). Near-miss titles are the most common source of noise — e.g. excluding "direct procurement" on a strategic-sourcing search. Stacks on top of the standing list, never replacing it. | | `removeAgency` | boolean | Exclude staffing agencies. Defaults to true; pass false to include. · default `true` | | `excludeAtsDuplicates` | boolean | Boards pool only: drop board rows the supplier already matched to an ATS posting. Defaults to true: the duplicate is filtered before it is paid for. · default `true` | | `description` | string | Boolean matched against the job description. ANDed with query — it NARROWS the title search rather than widening it. You can OMIT query entirely and sweep on description alone, which catches roles whose title matches nothing you would think to search for. Broader, so it returns more rows. | | `pool` | string | Which supply to ask: 'ats' (default — company career sites), 'boards' (job boards), or 'both'. These are different inventories, not two views of one, so 'both' searches twice and costs twice. · one of: `ats`, `boards`, `both` · default `ats` | | `forSomeoneElse` | boolean | Turn OFF the owner's exclusion lists and country guard; never scored | | `withinDays` | number | How far back to look. Searching is capped at 7 days — the supplier offers no fetch window between a week and six months. Counting reaches 30. The window used is always reported back as `window`. · default `7` · range 1–30 | | `remoteOnly` | boolean | Only roles that can be done remotely | | `limit` | number | Max postings per pool. The caller pays per row returned, so this is the spend dial for a one-off. Values above 100 are clamped to 100; `truncated` in the response says whether the ceiling cut the results. · default `10` · range 1–100 | **Example** ```json { "name": "search_jobs", "arguments": { "query": "\"VP of Engineering\" OR \"Head of Platform\"", "location": "Charlotte, NC", "seniority": [ "Director", "Executive" ], "experienceLevel": "5-10,10+", "excludeTitles": [ "recruiter", "talent acquisition" ], "remoteOnly": true, "limit": 10 } } ``` ## count_jobs How many postings match, without fetching any of them. Use this to size or debug a query before spending on it: it returns the same compiled expression, the same filter warnings and a total, but buys no rows. Iterate here first, then run search_jobs or create_search once the query looks right. | Argument | Type | Notes | |---|---|---| | `query` | string | Same syntax as search_jobs. | | `location` | string | e.g. "United States". Omit for worldwide, which counts far more. | | `seniority` | array of string | Job-board levels; career sites ignore them. · one of: `Internship`, `Entry level`, `Associate`, `Mid-Senior level`, `Director`, `Executive` | | `experienceLevel` | string | '0-2', '2-5', '5-10', '10+' — one, or several as a comma string. · one of: `0-2`, `2-5`, `5-10`, `10+`, `5-10,10+`, `2-5,5-10`, `0-2,2-5` | | `employmentType` | array of string | Work types. Defaults to FULL_TIME. · one of: `FULL_TIME`, `PART_TIME`, `CONTRACTOR`, `TEMPORARY`, `INTERN`, `VOLUNTEER`, `PER_DIEM`, `OTHER` | | `minEmployees` | number | Minimum organisation headcount. · range 1– | | `maxEmployees` | number | Maximum organisation headcount. · range 1– | | `industry` | array of string | Only these organisation industries. | | `companies` | array of string | Only these organisations, by name. | | `removeAgency` | boolean | Exclude staffing agencies. Defaults to true. · default `true` | | `description` | string | ANDed with query — it narrows. | | `remoteOnly` | boolean | Only roles that can be done remotely. | | `withinDays` | number | How far back to look. Searching is capped at 7 days — the supplier offers no fetch window between a week and six months. Counting reaches 30. The window used is always reported back as `window`. · default `7` · range 1–30 | | `pool` | string | 'ats' (default), 'boards', or 'both'. | **Example** ```json { "name": "count_jobs", "arguments": { "query": "\"Director Strategic Sourcing\"", "location": "United States", "withinDays": 30 } } ``` ## list_matches Scored postings from the owner's pipeline with score + written rationale. Filters: status, minScore, source. By default, matched postings scoring below their search's score threshold are omitted and COUNTED in hiddenBelowThreshold — when that count is non-zero, the list is not the whole pipeline. Pass includeBelowThreshold: true to see them, each marked belowThreshold. view: "pursuing" returns instead the Pursuing screen: every job between Pursue and Applied with its `stage` (queued → drafted → reviewed → final → applied, derived from the studio session and the documents), its one primary `action`, and its `artifacts`. | Argument | Type | Notes | |---|---|---| | `view` | string | The Pursuing screen (stage + one action per job in flight). · one of: `pursuing` | | `status` | string | Pipeline column to filter to. | | `minScore` | number | Only matches scoring at least this. · range 0–10 | | `source` | string | internal = found by Jobably’s own scans; ingest = saved with the browser extension. · one of: `ingest`, `internal` | | `includeBelowThreshold` | boolean | Also return matched postings scoring below their search's score threshold, marked belowThreshold: true. | | `sinceDays` | number | Only postings found within the last N days (1 = the last 24 hours). · range 1–365 | | `searchId` | string | Only postings found by this saved search. | | `limit` | number | Page size. · default `50` · range 1–200 | | `offset` | number | Rows to skip (paging). · default `0` · range 0– | **Example** ```json { "name": "list_matches", "arguments": { "status": "matched", "minScore": 7 } } ``` ## get_job Full posting detail: JD, score+rationale (with `flags.lean` — leader/doer/analyst shares of what the posting is buying, read from the responsibilities; it never moves the score), history, notes, artifacts. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | **Example** ```json { "name": "get_job", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" } } ``` ## set_job_status Move a posting through the pipeline (matched→pursuing→tailored→applied→screening→interviewing→offer→closed_*, dismissed↔matched). screening = a recruiter screen or phone call; it is optional (applied→interviewing is legal). Two axes, one column: closed_lost = rejected, closed_withdrawn = withdrawn, closed_won = accepted, and the stage reached before closing is read from the history (get_board returns it as `reached`). Illegal transitions are rejected by the server; closed stays closed. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `status` **(required)** | string | matched \| pursuing \| tailored \| applied \| screening \| interviewing \| offer \| closed_won \| closed_lost \| closed_withdrawn \| dismissed | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "set_job_status", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "status": "pursuing", "confirm": true } } ``` ## rate_job Rate a posting with a reason code AND a free-text note — the note tunes future scoring. Reasons: wrong_location, wrong_level, wrong_industry, comp_too_low, reputation, not_interested, good_fit, other. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `reasonCode` **(required)** | string | Why this rating. Feeds future scoring. · one of: `wrong_location`, `wrong_level`, `wrong_industry`, `comp_too_low`, `reputation`, `not_interested`, `good_fit`, `other` | | `note` | string | Free text. The note itself tunes future scoring — worth writing. | **Example** ```json { "name": "rate_job", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "reasonCode": "good_fit", "note": "Strong platform scope; comp in range." } } ``` ## tailor_resume Queue a ONE-SHOT provenance-guarded tailored resume for a posting: the whole document is rewritten in the background and the artifact lands in the app. Returns an operation id; poll get_operation. When the user wants to see and decide each change, use the STUDIO instead: open_tailoring_session → review_tailoring_session → get_tailoring_session (read the proposals) → decide_tailoring → build_tailored_resume. The studio and this tool produce the same artifacts under the same guards; the studio applies nothing the user did not accept. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "tailor_resume", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "confirm": true } } ``` ## revise_tailored Apply the author’s review decisions to the latest tailored draft (approve-and-decline). A finished tailor’s get_operation result carries `triage`: posting phrases the draft does not use, each verdict-sorted (supported / judgement / unsupported) with evidence from the record, a suggestion, and — when there is a real limit — a `caution` naming what must NOT be claimed (show it beside the item; it binds the revision too). Pass one decision per item id — approve or decline, optionally with the author’s own wording. Only supported/judgement items can be approved; unsupported items are the honest gap list and cannot be worked in. Returns an operation id; poll get_operation. The result lists `declined` (changes refused as untruthful — a correct outcome), `unauthorizedChanges` (model edits outside the approved scope, reverted), and `changedLines` (what actually changed). Every unchanged line ships byte-identical. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `decisions` **(required)** | array of object | One per triage item the author reviewed: {id, action: approve\|decline, wording?}. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "revise_tailored", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "decisions": [ { "id": "t1", "action": "approve", "wording": "Led supplier consolidation across 14 vendors." }, { "id": "t2", "action": "decline" } ], "confirm": true } } ``` ## list_tailoring_sessions Every tailoring-studio session on the account, newest first: job, status (new | reviewing | reviewed | built), master version, decided/proposal counts, and the built artifact id when there is one. | Argument | Type | Notes | |---|---|---| ## open_tailoring_session Open (or return) the tailoring-studio session for a job. Starts from the master version marked exemplar, else the latest; pass resumeId (from list_resumes) to start from another version — that REPLACES an existing session on a different version, discarding its proposals and decisions. Returns the session: sections as numbered lines (index, text, role), proposals (empty until reviewed), decisions, the job, and the master. Refuses (409) a job with no stored description or an account with no master. Next: review_tailoring_session. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `resumeId` | string | Optional master version id to start from (list_resumes → versions). | **Example** ```json { "name": "open_tailoring_session", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" } } ``` ## get_tailoring_session The full studio session for a job: `session.sections[]` each with `lines[]` ({index, text, role}) and `suggestions[]` ({id, kind: reword|add, lineIndex, original, proposed, reason, postingTerms, source, guard: {ok, reason}, styleNotes}) — a suggestion with guard.ok false was HELD BACK by the no-fabrication guards and cannot be accepted; show it with its reason. Also `session.analysis` (posting signal, verified terminology aliases, `headline.proposed`, `layout[]` recommendations — only a section-order one can be applied for the user), `session.decisions`, `session.lineOverrides` (an override with `budget: true` is an experience bullet the bullet budget left out for length when the session opened — `session.bulletBudget` carries the tally; the user can put one back with decide_tailoring `omit: false`), `session.build` after a build, `progress`, `activeOperation` while a review or build runs, and `latestOperation`. Read this after review_tailoring_session finishes, then walk the proposals WITH the user before calling decide_tailoring. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | **Example** ```json { "name": "get_tailoring_session", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" } } ``` ## review_tailoring_session Queue the studio review: one posting analysis (key terms, requirements, verified terminology aliases; on the user’s own model also layout recommendations and a headline), then every section reviewed line by line and role by role, each proposal checked against the record before it is shown. Returns an operation id; poll get_operation until done, then get_tailoring_session. Optional `sections` (indexes) re-reviews only those; `force` re-runs the posting analysis too. Refuses (409) while a review or build is already running. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `sections` | array of number | Section indexes to (re-)review. Omit for all. | | `force` | boolean | Re-run the posting analysis and layout advice as well. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "review_tailoring_session", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "confirm": true } } ``` ## decide_tailoring Save the user’s decisions on a reviewed session. `decisions`: one per proposal id — accept, decline, or edit with the user’s own `text`. `lineOverrides`: the user’s own wording (`text`) for ANY line, or `omit: true` to leave a line out of this draft (the master is never changed). `headline`: original | proposed | custom (+ text). `sectionOrder`: every section title exactly once, or null to restore the master’s order. `layoutDecisions`: accept/decline a recommendation id (accepting a section-order one applies it). Every piece of user wording is checked against the record as it lands — dates, figures, every number on the line, traceability — and anything refused comes back in `rejected` with the reason and is NOT saved; tell the user why. A proposal the user does not decide is not applied. Refuses (409) while a review is running. Returns the updated session and progress. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `decisions` | array of object | Proposal decisions: {id, action: accept\|decline\|edit, text?}. | | `lineOverrides` | array of object | Per-line overrides: {sectionIndex, lineIndex, text?, omit?}. | | `headline` | object | | | `sectionOrder` | array,null | Every section title exactly once, or null for the master’s order. | | `layoutDecisions` | array of object | Layout recommendation decisions: {id, action: accept\|decline}. | **Example** ```json { "name": "decide_tailoring", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "decisions": [ { "id": "s1", "action": "accept" }, { "id": "s2", "action": "edit", "text": "- Led supplier consolidation across 14 vendors, saving $4.8M." }, { "id": "s3", "action": "decline" } ], "lineOverrides": [ { "sectionIndex": 1, "lineIndex": 7, "omit": true } ], "headline": { "action": "proposed" }, "layoutDecisions": [ { "id": "l1", "action": "accept" } ] } } ``` ## build_tailored_resume Build the document from the session’s decisions: composed by code (undecided proposals are NOT applied), every changed line re-verified, rendered into the user’s own Word file when the master is a .docx, stored as the same artifacts a one-shot tailor produces (tailored resume + optional cover letter), scored, and the job moved to Tailored. Refuses (422) with `problems` naming any line that cannot be traced to the record — fix those with decide_tailoring first. Returns an operation id; poll get_operation (kind tailor_build); the artifact ids are in its result and get_tailored_document reads them back as text. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `coverLetter` | boolean | Also write a matching cover letter (default true). | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "build_tailored_resume", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "coverLetter": true, "confirm": true } } ``` ## discard_tailoring_session Start over: delete a job’s studio session (its proposals and decisions). Documents already built stay on the job. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "discard_tailoring_session", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "confirm": true } } ``` ## get_prep_packet Queue an interview prep packet (research brief, themes, STAR stories, draft-only outreach). | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "get_prep_packet", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "confirm": true } } ``` ## get_board Kanban board snapshot: cards per live status; applied cards carry days-since-applied aging (followUp.due at 5+ days with nothing back). Closed applications are NOT in the columns — `closed` lists them with `outcome` (rejected | withdrawn | accepted) and `reached` (the stage before closing: applied | screening | interviewing | offer), and `outcomes` is the tally (stillOpen, rejected, withdrawn, accepted, daysToHearBack, answeredShare, appliedEver). | Argument | Type | Notes | |---|---|---| ## move_card Move a board card to another column. Identical to set_job_status; provided so board-centric agents find it. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `toStatus` **(required)** | string | Destination column. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "move_card", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "toStatus": "applied", "confirm": true } } ``` ## list_searches Saved searches with stored query translation and warnings. | Argument | Type | Notes | |---|---|---| ## create_search Create a saved search that scans automatically from now on, scores every result against the owner's rubric, and applies their exclusion lists. This — not search_jobs — is the tool that carries the full filter set. Set location and a level filter unless the user genuinely wants worldwide, all-levels results — an unfiltered saved search spends money on every run for roles the candidate cannot take. COVERAGE: each run fetches up to 100 postings per source (career sites and job boards separately), and only those added since the previous run — so a saved search keeps up with a busy query over time rather than sampling it, which a single one-off search cannot do. Use count_jobs to size a query first: if it matches far more per run than the cap, narrow it or shorten the cadence. | Argument | Type | Notes | |---|---|---| | `name` **(required)** | string | What to call this search in the app. | | `query` | string | Boolean title search. A comma separates alternatives (reads as OR). | | `location` | string | e.g. "United States", "Charlotte, NC". Omitted means worldwide, which costs more. | | `seniority` | array of string | LinkedIn vocabulary: Internship, Entry level, Associate, Mid-Senior level, Director, Executive. Boards pool only; the ATS pool ignores it, so pair it with experienceLevel. · one of: `Internship`, `Entry level`, `Associate`, `Mid-Senior level`, `Director`, `Executive` | | `experienceLevel` | string | Years of experience: '0-2', '2-5', '5-10', '10+' — one, or several as an array or comma string ('5-10,10+'). The ONLY level filter the ATS pool honours; a saved search without it collects junior roles from career sites on every run. · one of: `0-2`, `2-5`, `5-10`, `10+`, `5-10,10+`, `2-5,5-10`, `0-2,2-5` | | `employmentType` | array of string | FULL_TIME, PART_TIME, CONTRACTOR, TEMPORARY, INTERN, VOLUNTEER, PER_DIEM, OTHER. Defaults to FULL_TIME. · one of: `FULL_TIME`, `PART_TIME`, `CONTRACTOR`, `TEMPORARY`, `INTERN`, `VOLUNTEER`, `PER_DIEM`, `OTHER` | | `minEmployees` | number | Minimum organisation headcount. · range 1– | | `maxEmployees` | number | Maximum organisation headcount. · range 1– | | `industry` | array of string | Organisation industries to include | | `companies` | array of string | Only these organisations, by name (filtered at the supplier). | | `excludeCompanies` | array of string | Drop these organisations. Stacks on top of the owner's standing lists. | | `excludeTitles` | array of string | Drop postings whose title contains any of these. Stacks on top of the owner's list. Direct-sourcing titles are the classic case. | | `removeAgency` | boolean | Exclude staffing agencies. Defaults to true. · default `true` | | `excludeAtsDuplicates` | boolean | Job-board pool only: drop board rows the supplier already matched to an ATS posting. Defaults to true: the duplicate is filtered before it is paid for, which matters on a search that runs on a schedule. · default `true` | | `description` | string | Boolean matched against the job description. ANDed with query — it NARROWS the search. Do not repeat the title terms here; that intersection is usually empty. Omit query entirely for a description-only sweep. One of query or description is required — a search with neither would fetch the whole market. | | `remoteOnly` | boolean | Only roles that can be done remotely. | | `scoreThreshold` | number | Only surface matches scoring at least this. · default `7` · range 0–10 | | `cadenceHours` | number | Hours between scans (default 6; 24 = daily). The response echoes the STORED value — verify it there. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "create_search", "arguments": { "name": "VP Engineering — remote, US", "query": "\"VP Engineering\" OR \"VP of Engineering\"", "location": "United States", "experienceLevel": "5-10,10+", "excludeTitles": [ "recruiter" ], "remoteOnly": true, "scoreThreshold": 7, "cadenceHours": 24, "confirm": true } } ``` ## pause_search Pause or resume a saved search. Pausing preserves the scan watermark; searches are never deleted. | Argument | Type | Notes | |---|---|---| | `searchId` **(required)** | string | Saved-search id, from list_searches. | | `active` **(required)** | boolean | true resumes, false pauses. The scan watermark survives either way. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "pause_search", "arguments": { "searchId": "a1f4d8e2-6b3c-4d97-9e05-7f8a9b0c1d2e", "active": false, "confirm": true } } ``` ## get_operation The outcome of something you queued — tailor_resume, get_prep_packet or a score. Those tools return an operation id and nothing else, so without this an agent can START work it can never observe: no way to tell finished from failed, and no way to read the result. Poll this until status is `done` or `error`. A finished TAILOR returns `unevidenced` — what the posting asks for that the record does not evidence. Show that list to the user: the draft never claims those things, and they are the fastest guide to what is worth adding to the accomplishment bank. A finished tailor also names its `renderer`: `into-master` means the .docx was rendered into the user’s own Word file (their fonts, margins, numbering); `mirrored` means the generic layout, with `rendererFallbackReason` saying why — tell the user when `masterIsDocx` is true, because their formatting was not applied. A finished tailor also carries `triage` (feed revise_tailored), `verbatimSections` with `verbatimReasons` (which sections kept the user’s own words and WHY — a deliberate skip, the time budget, or the failure; tell the user the reason, not just the name), the tally `tailoredSections` / `skippedSections` / `failedSections` with `degraded: true` whenever any section failed (a partly-tailored draft must never be presented as fully tailored; a run where nothing could be tailored is an `error`, retryable, not a `done`), `styleNotes` (advisory style observations, never blocking), `orientation` (what the posting is buying — leader / doer / analyst — vs what the draft reads as; order and emphasis only, a mismatch is a flag not a defect), and `strategy` (`single-pass` = one whole-resume call on the user’s own model; `chunked` = per-section, with `strategyFallbackReason` when a single pass fell back). A finished REVISE carries `declined`, `unauthorizedChanges`, and `changedLines` — show `declined` to the user: those are changes refused as untruthful, which is the guard working. A finished STUDIO REVIEW (kind tailor_review) carries the tally `reviewed` / `suggestions` / `heldBack` / `layout` and `layoutSkipped` when layout advice did not run; the proposals themselves are read with get_tailoring_session. A finished STUDIO BUILD (kind tailor_build) carries `artifactId`, `coverLetterArtifactId`, `changedLines` (the receipt of what the author accepted), `unevidenced`, `triage`, `scorecard`, `renderer`, and `studio: true`. | Argument | Type | Notes | |---|---|---| | `operationId` **(required)** | string | The id returned when the work was queued. | **Example** ```json { "name": "get_operation", "arguments": { "operationId": "c3d5e7f9-1a2b-4c6d-8e0f-2a3b4c5d6e7f" } } ``` ## list_accomplishments The accomplishment bank: the record tailoring draws on ALONGSIDE the master resume. A tailored resume may only use figures that appear here or in the master, verbatim — so this is also the list of figures that can legally appear in a draft. | Argument | Type | Notes | |---|---|---| ## add_accomplishments Add accomplishments to the bank. APPENDS — existing entries are never replaced, and duplicates are skipped, so this is safe to call repeatedly. Widening the bank is the single most effective way to improve tailored resumes: tailoring can only rearrange what it has been given, and a master resume is one compressed snapshot of a career. Keep figures EXACTLY as the user states them ($4.8M stays $4.8M) — a reworded number is unusable, because the provenance guard matches verbatim. One accomplishment per entry. The response reports what was added, skipped and dropped. For a whole LinkedIn history, the app takes the export zip directly under Profile. | Argument | Type | Notes | |---|---|---| | `bank` **(required)** | array of string | Accomplishments to add, ONE PER ARRAY ELEMENT, figures verbatim — e.g. ["Negotiated $4.8M savings, 20% on $24M of capital equipment spend.", "Cut sourcing cycle time 38% across a $40M program."]. Pass a real JSON array, not a quoted string. A newline-separated string is also accepted (one accomplishment per line). Commas NEVER separate entries — a comma is punctuation inside an accomplishment, so never pre-split on it. | **Example** ```json { "name": "add_accomplishments", "arguments": { "bank": [ "Negotiated $4.8M savings, 20% on $24M of capital equipment spend.", "Cut sourcing cycle time 38% across a $40M program." ] } } ``` ## update_search Edit a saved search: its query, cadence, threshold, or any filter. Send ONLY what you want to change — every filter you do not mention is preserved. (The underlying API replaces the filter object wholesale, so this tool reads the stored search and merges; a client that rebuilt that object from its own inputs would silently DELETE every filter it did not know about.) Use pause_search to stop or resume one; this is for changing what it seeks. | Argument | Type | Notes | |---|---|---| | `searchId` **(required)** | string | Saved-search id, from list_searches. | | `name` | string | Rename the search. | | `query` | string | Boolean title search. Replaces the existing query. | | `location` | string | e.g. 'United States'. Empty string means worldwide. | | `description` | string | Boolean matched against the description. NARROWS. | | `experienceLevel` | string | '0-2', '2-5', '5-10', '10+', or a comma list. | | `seniority` | array of string | Job-board levels. | | `employmentType` | array of string | Work types. | | `industry` | array of string | Organisation industries. | | `companies` | array of string | Only these organisations. | | `excludeCompanies` | array of string | Drop these organisations. | | `excludeTitles` | array of string | Drop these title substrings. | | `minEmployees` | number | Minimum headcount. · range 1– | | `maxEmployees` | number | Maximum headcount. · range 1– | | `remoteOnly` | boolean | Only roles that can be done remotely. | | `removeAgency` | boolean | Exclude staffing agencies. | | `scoreThreshold` | number | Surface floor. · range 0–10 | | `cadenceHours` | number | Hours between scans. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "update_search", "arguments": { "searchId": "a1f4d8e2-6b3c-4d97-9e05-7f8a9b0c1d2e", "location": "United States", "excludeTitles": [ "recruiter", "staffing" ], "confirm": true } } ``` ## run_search_now Run a saved search immediately instead of waiting for its cadence. Optionally reach further back than the watermark with backfillHours (max 168 = 7 days) to pick up postings from before the search existed. A paused search is refused — resume it first. | Argument | Type | Notes | |---|---|---| | `searchId` **(required)** | string | Saved-search id, from list_searches. | | `backfillHours` | number | Look back this many hours instead of since the last run. Max 168. · range 1–168 | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "run_search_now", "arguments": { "searchId": "a1f4d8e2-6b3c-4d97-9e05-7f8a9b0c1d2e", "backfillHours": 48, "confirm": true } } ``` ## get_search_health Why is a search returning nothing? Per search and per source: the last run, how many postings came back, how many were new, whether the run failed and why, and the COMPILED query the supplier actually received. This is the tool for diagnosing a quiet search — an empty result and an impossible query look identical from outside, and the compiled expression is what tells them apart. | Argument | Type | Notes | |---|---|---| ## rescore_job Re-score ONE posting against the current rubric. Use after the questionnaire or exclusion lists change, or when a score looks wrong. Returns an operation id — poll get_operation. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "rescore_job", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "confirm": true } } ``` ## add_job_note Attach a note to a posting — a recruiter conversation, a referral, why it was interesting. Notes are first-class: they are what the owner reads later, when the score has stopped being the interesting part. | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | | `note` **(required)** | string | The note text. | **Example** ```json { "name": "add_job_note", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b", "note": "Recruiter called — onsite loop scheduled for Tuesday." } } ``` ## declare_applied Record an application made OUTSIDE Jobably — through a recruiter, a referral, or a company site. Creates the board card if the posting is not already tracked. Idempotent: declaring the same role twice does not duplicate it. | Argument | Type | Notes | |---|---|---| | `company` **(required)** | string | Employer name. | | `title` **(required)** | string | Role title. | | `url` | string | Posting URL, if known. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "declare_applied", "arguments": { "company": "Globex", "title": "VP Operations", "url": "https://boards.example.com/987", "confirm": true } } ``` ## get_exclusions The two standing exclusion lists: `fit` (poor-fit employers, which the rubric could in principle learn) and `reputation` (vetoes that apply regardless of role). These shape every search and every score, so read them before concluding a search is behaving oddly. | Argument | Type | Notes | |---|---|---| ## set_exclusions REPLACES both exclusion lists. Call get_exclusions first and send the complete lists including everything being kept — a partial call silently drops the rest. The two are different: `fit` is about match quality, `reputation` is a veto that applies whatever the role. | Argument | Type | Notes | |---|---|---| | `fit` **(required)** | array of string | Poor-fit employers. Complete list. | | `reputation` **(required)** | array of string | Reputation vetoes, independent of role. Complete list. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "set_exclusions", "arguments": { "fit": [ "Acme Staffing" ], "reputation": [ "BadCo Industries" ], "confirm": true } } ``` ## list_resumes Master resume versions and every tailored artifact produced so far, with the job each belongs to. Use it to tell the owner what already exists before spending tailor quota producing it again. | Argument | Type | Notes | |---|---|---| ## list_operations Everything queued or recently finished — tailoring, prep packets, scoring — with status and any failure reason. Check here before starting metered work: what the user asked for may already be running. | Argument | Type | Notes | |---|---|---| ## get_tailored_document READ a tailored resume, cover letter or prep packet as text. list_resumes gives you the artifact ids. This is what makes a draft reviewable: without it you can hand the owner a download link but cannot see what was produced, so you can neither critique it nor help them check it. ALWAYS read the draft before telling anyone it is ready — and read the gap list from get_operation alongside it. The guards stop the draft inventing figures; they cannot stop it being badly aimed, or carrying a credential from another industry onto this application. | Argument | Type | Notes | |---|---|---| | `artifactId` **(required)** | string | Artifact id, from list_resumes or get_job. | **Example** ```json { "name": "get_tailored_document", "arguments": { "artifactId": "d4e6f8a0-2b3c-4d5e-9f01-3c4d5e6f7a8b" } } ``` ## get_master_resume The parsed master resume: name, headline, contact and every section, as stored. This is the record tailoring mirrors — sections come out in this order, under these titles — so read it to understand why a draft is shaped the way it is, and to see what the record does NOT contain before promising it can. | Argument | Type | Notes | |---|---|---| ## mark_exemplar PIN a master resume version for the tailoring studio, or unpin it: a fresh studio session starts from the newest pinned version instead of the newest upload. It does NOT teach tailoring a style (that is a sent final marked as an exemplar, in the app’s Drafts). Warn the owner when the pinned version is older than their newest master — the studio keeps reading the old one. | Argument | Type | Notes | |---|---|---| | `resumeId` **(required)** | string | Resume version id, from list_resumes. | | `isExemplar` **(required)** | boolean | true pins it for the studio, false unpins it. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "mark_exemplar", "arguments": { "resumeId": "e5f7a9b1-3c4d-4e6f-a012-4d5e6f7a8b9c", "isExemplar": true, "confirm": true } } ``` ## delete_tailored_document Delete a tailored artifact. The posting, its score and its history are untouched — only the generated document goes. Use it to clear a draft the owner rejected so the next one is unambiguous. | Argument | Type | Notes | |---|---|---| | `artifactId` **(required)** | string | Artifact id, from list_resumes. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "delete_tailored_document", "arguments": { "artifactId": "d4e6f8a0-2b3c-4d5e-9f01-3c4d5e6f7a8b", "confirm": true } } ``` ## get_usage What has this cost me? Cycle usage vs plan quotas + estimated spend. | Argument | Type | Notes | |---|---|---| ## get_shortlist TODAY'S SHORT LIST — the jobs worth the owner's time today, best first. Start every daily check-in here. Built each morning from new scored matches that clear the owner's bar (the same bar Matches uses), held stable for the day, and returned LIVE: each item carries its current status, score and written rationale, its stage if being pursued, any tailored documents, `appUrl` (the job in Jobably) and `reviewUrl` (where a draft is reviewed and downloaded). `carried: true` marks a match from an earlier list the owner has not acted on. `criteria.awaitingScore` counts recent matches not scored yet — they cannot make the list until they are, so an empty list with a non-zero awaitingScore means "scoring in progress", not "nothing fits". Pass `day` (YYYY-MM-DD) to read a past list. | Argument | Type | Notes | |---|---|---| | `day` | string | A past UTC day, YYYY-MM-DD. Omit for today. | ## regenerate_shortlist Rebuild TODAY'S short list from the current matches — after the owner changed their floor or exclusions, or new matches were scored since this morning. Replaces the list everyone sees today. | Argument | Type | Notes | |---|---|---| | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "regenerate_shortlist", "arguments": { "confirm": true } } ``` ## list_applications Everything the owner has applied to, newest first: status, date applied, whether a follow-up is due, the documents (including the final they sent), `appUrl` and `reviewUrl`. Record a new application with move_card (toStatus "applied") for a tracked job, or declare_applied for one found elsewhere. | Argument | Type | Notes | |---|---|---| | `sinceDays` | number | Applied within N days. · range 1–730 | | `status` | array of string | Only these statuses, e.g. ["applied","screening","interviewing"]. | **Example** ```json { "name": "list_applications", "arguments": { "sinceDays": 30, "status": [ "applied", "screening", "interviewing" ] } } ``` ## show_tailored_resume Show what the tailored resume for a job LOOKS like: the newest draft as text, its version, every other document for the job (cover letter, prep packet, the final the owner sent), and `reviewUrl` — where the owner reviews, finalizes and downloads it. Present the text to the owner; link reviewUrl for the formatted document. If no draft exists yet, says so (tailor_resume or the studio tools make one). | Argument | Type | Notes | |---|---|---| | `jobId` **(required)** | string | Job id, from list_matches or get_board. | **Example** ```json { "name": "show_tailored_resume", "arguments": { "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" } } ``` ## query_jobs The flexible read for building the owner a table, report or list: any mix of statuses, a score range, found within N days, applied within N days, one search, one list, a company or a title/company text match, sorted by score, newest found or newest applied. Light columns (title, company, location, status, score, rationale, dates, appUrl) — use get_job for a full posting. | Argument | Type | Notes | |---|---|---| | `status` | array of string | Any of these pipeline statuses. | | `minScore` | number | range 0–10 | | `maxScore` | number | range 0–10 | | `sinceDays` | number | Found within N days. · range 1–365 | | `appliedSinceDays` | number | Applied within N days. · range 1–730 | | `searchId` | string | Saved-search id, from list_searches. | | `listId` | string | Only jobs on this list (list_lists). | | `company` | string | Company name contains. | | `q` | string | Title or company contains. | | `sort` | string | one of: `score`, `found`, `applied` · default `found` | | `limit` | number | default `50` · range 1–200 | | `offset` | number | default `0` · range 0– | **Example** ```json { "name": "query_jobs", "arguments": { "status": [ "matched", "pursuing" ], "minScore": 8, "sinceDays": 7, "sort": "score", "limit": 20 } } ``` ## get_insights The numbers for a report or dashboard: jobs by status, the score histogram, found and applied per week (12 weeks), the funnel from seen to closed, hit rate by search, median days from found to applied, open applications by age, and which agents acted on the account in the last 30 days. | Argument | Type | Notes | |---|---|---| ## get_activity The activity record: who — the owner in the app, a personal access token, or a connected agent like you — did what, newest first, with the tool, the target and the outcome. Every change to the account is recorded before it is made. Filter by `via` (session | pat | oauth) or `client` (an agent or token name). Page with `before` set to the previous page’s nextBefore. | Argument | Type | Notes | |---|---|---| | `via` | string | one of: `session`, `pat`, `oauth`, `system` | | `client` | string | Agent or token name, e.g. "ChatGPT". | | `sinceDays` | number | range 1–3650 | | `before` | string | ISO timestamp cursor from nextBefore. | | `limit` | number | default `50` · range 1–200 | **Example** ```json { "name": "get_activity", "arguments": { "via": "oauth", "limit": 20 } } ``` ## list_lists The owner’s named job lists ("Dream companies", "Apply this week") with item counts. Lists are where an agent keeps a working set for the owner; they appear in the app and can be embedded in a report. | Argument | Type | Notes | |---|---|---| ## create_list Create a named list of jobs for the owner. Add jobs with add_to_list. Name it the way the owner would. | Argument | Type | Notes | |---|---|---| | `name` **(required)** | string | List name, up to 120 characters. | | `description` | string | What the list is for. | **Example** ```json { "name": "create_list", "arguments": { "name": "Apply this week", "description": "Top picks from the short list." } } ``` ## get_list One list with its jobs: title, company, status, score, note, appUrl. | Argument | Type | Notes | |---|---|---| | `listId` **(required)** | string | List id, from list_lists. | **Example** ```json { "name": "get_list", "arguments": { "listId": "f6a8b0c2-4d5e-4f70-b123-5e6f7a8b9c0d" } } ``` ## update_list Rename a list or change its description. | Argument | Type | Notes | |---|---|---| | `listId` **(required)** | string | List id, from list_lists. | | `name` | string | New name. | | `description` | string | New description. | **Example** ```json { "name": "update_list", "arguments": { "listId": "f6a8b0c2-4d5e-4f70-b123-5e6f7a8b9c0d", "name": "Apply by Friday" } } ``` ## delete_list Delete a list. The jobs on it are untouched — only the list goes. | Argument | Type | Notes | |---|---|---| | `listId` **(required)** | string | List id, from list_lists. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "delete_list", "arguments": { "listId": "f6a8b0c2-4d5e-4f70-b123-5e6f7a8b9c0d", "confirm": true } } ``` ## add_to_list Add jobs to a list, optionally with a note (why they are there). Adding a job already on the list just updates its note. | Argument | Type | Notes | |---|---|---| | `listId` **(required)** | string | List id, from list_lists. | | `jobIds` **(required)** | array of string | Job ids, from get_shortlist, list_matches or query_jobs. | | `note` | string | Why these jobs are on the list. | **Example** ```json { "name": "add_to_list", "arguments": { "listId": "f6a8b0c2-4d5e-4f70-b123-5e6f7a8b9c0d", "jobIds": [ "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" ], "note": "Strong platform scope; comp in range." } } ``` ## remove_from_list Take one job off a list. The job itself is untouched. | Argument | Type | Notes | |---|---|---| | `listId` **(required)** | string | List id, from list_lists. | | `jobId` **(required)** | string | Job id, from list_matches or get_board. | **Example** ```json { "name": "remove_from_list", "arguments": { "listId": "f6a8b0c2-4d5e-4f70-b123-5e6f7a8b9c0d", "jobId": "b7e2c4d1-9f3a-4e58-8a06-1c2d3e4f5a6b" } } ``` ## list_reports Reports and dashboards saved to the owner’s account — by the owner or by an agent — with their widget types and `appUrl` (where the owner opens it in Jobably). | Argument | Type | Notes | |---|---|---| ## get_report One saved report. With render (the default) every widget comes back with its LIVE data — the same numbers the owner sees when they open it. | Argument | Type | Notes | |---|---|---| | `reportId` **(required)** | string | Report id, from list_reports or save_report. | | `render` | boolean | Resolve each widget to live data. · default `true` | **Example** ```json { "name": "get_report", "arguments": { "reportId": "a7b9c1d3-5e6f-4a81-8c34-6f7a8b9c0d1e" } } ``` ## save_report Save a report or dashboard to the owner’s account, where it stays LIVE: widgets are queries, not snapshots, so it is still true next week. Pass `reportId` to replace an existing one. Widget types: markdown {body} (your written analysis — headings, lists, tables, links); stats {} (insights); funnel {}; jobs {query: the query_jobs fields, limit ≤ 50}; list {listId}; shortlist {}; activity {via?, limit?}. Each takes an optional title. Up to 20 widgets. Returns `appUrl` — give it to the owner. | Argument | Type | Notes | |---|---|---| | `reportId` | string | Replace this report. Omit to create a new one. | | `title` **(required)** | string | Report title. | | `description` | string | One line on what it shows. | | `pinned` | boolean | Pin to the top of Reports. | | `widgets` **(required)** | array of object | | **Example** ```json { "name": "save_report", "arguments": { "title": "Monday pipeline review", "pinned": true, "widgets": [ { "type": "markdown", "body": "## This week\n- Two applications out, one screen booked." }, { "type": "shortlist", "title": "Today" }, { "type": "jobs", "title": "Strong matches not yet pursued", "query": { "status": [ "matched" ], "minScore": 8, "sort": "score", "limit": 10 } }, { "type": "funnel" } ] } } ``` ## delete_report Delete a saved report. | Argument | Type | Notes | |---|---|---| | `reportId` **(required)** | string | Report id, from list_reports. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "delete_report", "arguments": { "reportId": "a7b9c1d3-5e6f-4a81-8c34-6f7a8b9c0d1e", "confirm": true } } ``` ## get_profile The owner’s profile: the questionnaire their scoring rubric is built from (seniority floor, locations, arrangements, comp floor, industries, title keywords), notification preferences, and whether a master resume exists. Read it before set_questionnaire or set_preferences. Includes the interview answers and conditional rules set_questionnaire can merge. | Argument | Type | Notes | |---|---|---| ## set_questionnaire Set what the owner wants — the questionnaire every posting is scored against. MERGES: send only the fields that change; everything else is kept (interview answers merge field by field, conditional rules by id). Scoring needs this AND a master resume. Ask the owner rather than guessing — a wrong floor hides good jobs. | Argument | Type | Notes | |---|---|---| | `seniorityFloor` | string | Lowest level worth seeing, e.g. "Director". | | `locations` | array of string | Preferred locations (a ranking signal, not a filter). | | `arrangements` | array of string | Acceptable work arrangements. · one of: `remote`, `hybrid`, `onsite` | | `compFloor` | number | Annual base floor. | | `compTiers` | array of object | Pay tiers judged on the TOP of the published range, e.g. [{below: 200000, cap: 6}, {below: 175000, cap: 4}]: a range topping out below 200K scores at most 6. Replaces the floor line when set; [] clears. | | `engagementTypes` | array of string | Salaried, contract, or both. · one of: `w2`, `contract` | | `minHourlyRate` | number | Contract rate floor, $/hour. | | `industries` | array of string | Industries to favour. | | `excludedIndustries` | array of string | Industries to avoid. | | `titleKeywords` | string | Target titles, e.g. "VP Engineering" OR "Head of Platform". | | `excludedTitleKeywords` | array of string | Titles that disqualify a posting. | | `interview` | object | The owner’s own words, quoted verbatim into the rubric — the strongest lever there is. MERGED field by field: send only the answers that change. | | `conditionalRules` | array of object | Rules that apply only when their conditions hold, e.g. IC titles outside Charlotte score at most 5 (title contains any of [Manager, Analyst] AND location matches none of [Charlotte]). MERGED by id: a rule with an existing id replaces it. | | `removeConditionalRules` | array of string | Ids of conditional rules to delete. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "set_questionnaire", "arguments": { "seniorityFloor": "Director", "arrangements": [ "remote", "hybrid" ], "titleKeywords": "\"VP Engineering\" OR \"Head of Platform\"", "interview": { "compNotes": "Judge the top of the range: $200K+ no cap; $175–199K at most 6." }, "conditionalRules": [ { "id": "ic-charlotte", "description": "IC roles are in range only in the Charlotte metro", "condition": { "field": "location", "includesAny": [ "Charlotte" ] }, "effect": "an individual-contributor role is not capped for scope" } ], "confirm": true } } ``` ## set_preferences Notification and display preferences. MERGES over the stored ones. matchAlertFrequency: instant | daily | off (emails when a match clears alertThreshold); digestFrequency: daily | weekly | off; matchesMinScore: one floor for every search on Matches and the short list (useSearchThresholds: true goes back to each search’s own threshold); resumeBulletBudget: an overall cap on experience bullets in a tailored resume, on top of the per-role limits (current role 12, recent roles 6, older fewer); 0 keeps every bullet; useDefaultBulletBudget: true goes back to the per-role default. | Argument | Type | Notes | |---|---|---| | `matchAlertFrequency` | string | one of: `instant`, `daily`, `off` | | `digestFrequency` | string | one of: `daily`, `weekly`, `off` | | `alertThreshold` | number | Score that triggers a match email. · range 5–10 | | `matchesMinScore` | number | One floor for every search, on Matches and the short list. · range 0–10 | | `useSearchThresholds` | boolean | true clears matchesMinScore: each search uses its own threshold again. | | `resumeBulletBudget` | number | Overall cap on experience bullets, on top of the per-role limits (each role keeps at least one; what is left out is listed beside the draft). 0 keeps every bullet. · range 0–60 | | `useDefaultBulletBudget` | boolean | true clears resumeBulletBudget: Jobably’s default length applies again. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "set_preferences", "arguments": { "matchAlertFrequency": "daily", "matchesMinScore": 7, "confirm": true } } ``` ## upload_master_resume Upload the owner’s master resume — what every score and every tailored draft is built from. Send `text` (the resume as plain text or Markdown, exactly as written — never rewritten or summarized) or `fileBase64` + `filename` for a .docx or .pdf. A Word file keeps its formatting in tailored drafts; text does not, so prefer the file when the owner has one. Creates a new master version; earlier ones are kept. | Argument | Type | Notes | |---|---|---| | `text` | string | The resume, verbatim, as plain text or Markdown. | | `fileBase64` | string | A .docx or .pdf, base64-encoded. | | `filename` | string | File name with extension, e.g. "resume.docx". | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "upload_master_resume", "arguments": { "text": "# Jordan Lee\nVP Engineering\n\n## Experience\n**Acme** — VP Engineering, 2020 – Present\n- Led a 60-person platform org.", "confirm": true } } ``` ## get_started START HERE on a new account, or whenever the owner says "set me up" / "find me jobs". Returns where the account stands (`status`: resume, profile, searches, first scan, matches, scored), the NEXT step in words (`next.step`, `next.say`), and — once a master resume is uploaded and until the account is set up — a `proposal` drawn from it: a profile (`questionnaire`) and a first search, every value copied from the resume with its source in `basis`, plus `ask`: the questions only the owner can answer (pay floor, remote/hybrid/on-site, industries). Show the proposal, ask those questions, fold the answers into it, then call apply_setup. Steps: upload_resume → upload_master_resume; apply_setup → apply_setup; first_scan / scoring → wait a few minutes and call this again; score_matches → score_new_matches; ready → get_shortlist. | Argument | Type | Notes | |---|---|---| ## apply_setup Set the owner up in ONE call, after they approved the proposal from get_started: saves the profile (`questionnaire`, which builds the scoring rubric) and creates each search, and the first scan runs immediately. Matches arrive within minutes; on a new account they are scored automatically (`scoring: "automatic"`), otherwise call score_new_matches. Send the proposal back with the owner’s answers folded in — `questionnaire` is the WHOLE profile (seniorityFloor required), `searches` is a list of {name, query, filters, scoreThreshold, cadenceHours} exactly as the proposal shapes them (filters takes the create_search filter names, e.g. location, experienceLevel, remoteOnly). Every search is checked before anything is saved: one that could never match refuses the whole call, naming it. Needs a master resume (upload_master_resume). | Argument | Type | Notes | |---|---|---| | `questionnaire` **(required)** | object | The whole profile: seniorityFloor (required), titleKeywords, locations, arrangements (remote \| hybrid \| onsite), compFloor, engagementTypes (w2 \| contract), minHourlyRate, industries, excludedIndustries, excludedTitleKeywords. | | `searches` **(required)** | array of object | One to five searches, shaped like proposal.searches: {name, query, filters, sources?, scoreThreshold?, cadenceHours?}. | | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "apply_setup", "arguments": { "questionnaire": { "seniorityFloor": "VP", "titleKeywords": "\"VP Engineering\" OR \"Head of Platform\"", "locations": [ "Charlotte, NC" ], "arrangements": [ "remote", "hybrid" ], "compFloor": 250000, "engagementTypes": [ "w2", "contract" ] }, "searches": [ { "name": "VP Engineering roles", "query": "\"VP Engineering\" OR \"Head of Platform\"", "filters": { "location": "United States", "experienceLevel": "10+" }, "scoreThreshold": 7, "cadenceHours": 6 } ], "confirm": true } } ``` ## score_new_matches Score the owner’s unscored matches now, newest first. Subscribers’ new matches are scored automatically and so are a new account’s first; use this when get_started says `score_matches`. One run per account: asking again while one runs returns that run. Poll it with get_operation; scores appear on list_matches as they land. | Argument | Type | Notes | |---|---|---| | `confirm` **(required)** | boolean | Must be true. Required because this spends quota or changes state — ask the user first. | **Example** ```json { "name": "score_new_matches", "arguments": { "confirm": true } } ``` ## Resources | URI | Contents | |---|---| | `jobably://resume/master` | Latest master resume, parsed structure | | `jobably://board` | Current board snapshot | ## Errors Tool failures come back as tool results with `isError: true` and a readable message (quota exceeded, illegal transition, missing confirm) — read them rather than retrying blindly. Protocol-level errors (unknown tool, bad auth) are JSON-RPC errors. ## Your agent's workspace https://docs.jobably.com/mcp/workspace/ > The daily short list, applications, tailored drafts, lists, reports and the activity record — what your agent works with, and where you see it. Once your agent is [connected](/mcp/connect/), it works inside your Jobably account. Everything it makes stays there, so you see the same thing in the app that your agent sees. ## The daily short list Every morning Jobably builds a short list: the best new matches that clear your bar, best first. The bar is the same one the Matches page uses — your own floor if you set one in Settings, otherwise each search's score threshold — and a job you liked on Swipe makes the list even if it scored below it. - The list holds still for the day, so "the list" means the same jobs to you, your agent and the app. - A job has to be scored before it can make the list. If new matches are still being scored, the list says how many are waiting, and an empty list fills in as their scores land — it is not held empty for the day. - Each job shows its score and the written reason for it, where it stands (new, pursuing, drafted, applied), and any documents made for it. - A good match you have not acted on yet carries over to the next day's list, marked as coming from an earlier list. - Ask your agent to rebuild it after you change your floor or exclusions. Ask: *"What's on my short list today?"* · In the app: **Short list**. ## Links to your jobs Whenever your agent names a job, it can give you a link to it in Jobably. The link opens **Matches with that job first, expanded and scrolled into view** — however many matches you have and whatever the page is filtered to, including a job you already applied to or passed on. Dismiss or pursue it there and the next job takes its place. If a link says the job is not on this account, you are signed in to a different Jobably account from the one your agent is connected to. ## Tailored resumes, shown to you Your agent can tailor a resume for any job and **show you the draft as text** right in the conversation. Every draft is built only from your own history — no figure is ever invented, whatever the agent asks for. To see the formatted document and download it, your agent gives you a link to the review page in Jobably. The download is there, after you confirm you have reviewed it — an agent cannot skip that step for you. Ask: *"Show me the tailored resume for the Acme job."* ## Recording applications Jobably never applies to a job for you. When you have applied, tell your agent and it moves the job to **Applied** — or, for a role you found elsewhere, records it as applied so it joins your board. Your agent can list everything you have applied to, when, and which ones are due a follow-up. Ask: *"I applied to the Acme role this morning — mark it applied."* · *"What have I applied to this month, and who should I follow up with?"* ## Lists Lists are named sets of jobs your agent keeps for you — "Dream companies", "Apply this week" — each job with an optional note on why it is there. They can be embedded in a report. Ask: *"Make a list called Apply this week with the top three from today's short list."* ## Reports and dashboards Your agent can build you a report and save it to your account. A report is a page of parts — written notes, today's short list, a table of jobs, a list, your pipeline at a glance, the funnel from postings seen to applied, recent activity — and every number is read fresh each time you open it. A report your agent built on Monday is still true on Friday. Ask: *"Build me a weekly pipeline review: what's new, what I applied to, and where things stand."* · In the app: **Reports**. ## The activity record Every change to your account — by you, a token or a connected agent — is recorded before it happens, with who did it, which tool they used, what they acted on and how it turned out. Everything a token or an agent reads is recorded too. If the record cannot be written, the change is not made. See it in **Settings → Integrations → Activity**, filtered by who did it, or ask: *"What has my agent done on my account this week?"* ## Setting up through your agent A new account can be set up from the conversation: your agent can upload your master resume (paste it, or attach a Word or PDF file), fill in what you are looking for, and set your notification preferences. A Word file keeps its formatting in tailored drafts; pasted text does not, so attach the file if you have it. ## API overview https://docs.jobably.com/api/overview/ > The REST API behind everything — OpenAPI spec, auth, and conventions. Everything in Jobably — the app, the extension, the MCP server — drives one REST API at `api.jobably.com`. - **Interactive reference:** [api.jobably.com/docs](https://api.jobably.com/docs) - **OpenAPI 3.1 spec:** [api.jobably.com/openapi.json](https://api.jobably.com/openapi.json) :::caution[The OpenAPI document is not yet the whole API] It currently describes the meta endpoints (`/v1/version`, `/v1/status`) only. Every other route works and is stable — it is the machine-readable *description* that is incomplete, not the API. If you are building an agent, the [MCP tool reference](/mcp/tools/) is the complete machine-readable surface: it is generated from the running server, covers every capability, and documents every argument. If you are calling REST directly, this page and the guides are authoritative until the spec catches up. ::: ## Auth - **Session cookie** — issued at sign-in; used by the app. - **Personal access token** — `Authorization: Bearer jbly_pat_…`, scoped (`read`, `write`, `ingest`), created and revoked in app Settings. Token management itself is session-only: a stolen token cannot mint more tokens, plant provider keys, or delete the account. ## Conventions - Versioned under `/v1`; JSON everywhere. - Errors are [RFC 9457](https://www.rfc-editor.org/rfc/rfc9457) `application/problem+json` with a stable `type` URI — branch on `status` and `type`, not message strings. - Async AI work (tailor, revise, prep, scoring) returns `202` with an operation id; poll `/v1/operations/:id` or stream `/v1/operations/:id/events` (SSE). ## Typed clients The app's own client is generated from the served spec with `openapi-typescript` — the same spec is available to you at the URL above. ## Bring your own keys https://docs.jobably.com/byok/keys/ > Run resume reading, scoring, tailoring, and prep on your own model provider keys. Add an **Anthropic, OpenAI, Google, xAI, or Ollama Cloud** API key in app Settings and Jobably's AI operations run against your key instead of platform quotas. ## Keys are checked before they're saved When you add a key, Jobably first asks the provider to confirm it works. A key the provider rejects is never stored — you'll see a clear message so you can fix a typo or generate a fresh key. If the provider can't be reached at that moment, the key is saved anyway and simply used on your next operation. ## How keys are stored Each key is sealed with bank-grade encryption the moment it arrives, and keys are **write-only**: Jobably can use a key to run your work, show you its last four characters so you know which one it is, or delete it — but nothing, including our own screens and logs, can ever display a key back. Automated tests verify this on every release. ## Scope of use Your keys are used solely to run *your* workloads — reading the master resume you upload, scoring your postings, tailoring your resumes, building your prep packets. Costs accrue on your provider account under your agreement with them. If your provider rejects a saved key, Jobably stops and asks you to fix it rather than quietly switching to the included model. ## Removing a key Instant, from Settings. In-flight work falls back to platform models within your plan quotas. ## Pick your tailoring model With a key stored, Settings → Preferences → **Resume tailoring model** lets you run tailoring (and its cover letter) on a stronger model of your choice: - **Anthropic** — Claude Opus 5, Claude Sonnet 5 - **OpenAI** — GPT-5 - **Google** — Gemini 2.5 Pro - **xAI** — Grok 4.5, Grok 4.3 - **Ollama Cloud** — DeepSeek V4 Flash, Qwen 3.5, Kimi K3, GPT-OSS 120B Stronger models produce noticeably stronger resumes; the AI usage runs on your key (Ollama Cloud bills through your Ollama subscription rather than per token). If the key is removed, tailoring falls back to the platform default automatically. ## Billing & trial https://docs.jobably.com/billing/plan/ > The plan, the trial, quotas, and what happens when payment fails. Start free, no card required; two paid plans, each with a 7-day fully-featured free trial. The trial starts at checkout, which asks for a card: cancel before day seven and nothing is charged. Checkout and the billing portal run on Stripe — card details never touch Jobably. - **Free.** Save postings with the Chrome extension, build your scoring rubric, score up to 4 matches a day against it, and track everything on the pipeline board. Your first search scans and scores automatically so you see scored matches before deciding anything. - **Jobably Pro — $30/month.** The full system: scans every 6 hours, scoring with written rationales, tailored resumes with cover letters, prep packets, the pipeline board, extension, and MCP access. - **Jobably Executive — $100/month.** Everything in Pro, run harder: **hourly scans**, **new matches scored automatically** as they arrive, roughly **5× the monthly credits**, and up to 25 active searches. Built for Director-to-C-suite searches where timing decides. Switch between plans any time from Billing → Manage subscription. ## Credits (quotas) Each plan includes per-cycle credit pools for scores, tailored resumes, prep packets, Finder pulls, and jobs fetched. The Dashboard and Billing pages show live remaining-credit meters; the API returns `402` with reset guidance when a pool is exhausted. [BYOK](/byok/keys/) usage runs on your keys instead of the AI pools. ## Payment failure A failed renewal starts a **3-day grace period** with full access. If payment still fails, the account drops to **read-only**: your pipeline and artifacts stay readable and exportable, scans and AI operations pause. Paying again restores everything — nothing is deleted. ## Cancelling Billing → Manage subscription opens the Stripe portal. Cancellation applies at period end; read access continues; deleting the account (Settings) permanently erases everything with a verified cascade. ## FAQ https://docs.jobably.com/faq/ > Short answers, with links to the long ones. **Does Jobably apply to jobs for me?** No — deliberately. It discovers, scores, tailors, and tracks. Outreach materials are drafts; you send them. Automated applying is how accounts get flagged and reputations burned. **Can the tailored resume exaggerate?** It structurally cannot invent figures — see [provenance rules](/resumes/provenance/). Emphasis changes; facts don't. **Why did my search return nothing?** Check what the query compiled to in scan health — usually a [query trap](/searches/query-syntax/#traps-jobably-catches-for-you) like an over-tight phrase. **Is my data used to train models?** No. Your data serves you and nothing else. Deleting your account erases it with a verified cascade. **What can a stolen PAT do?** Only what its scopes allow — and never manage tokens, provider keys, or account deletion; those are session-only. Revoke tokens in Settings. **Where's the extension?** Live on the [Chrome Web Store](https://chromewebstore.google.com/detail/ehefnaajjdncohccmlgemcplonmjoffj) — see the [install docs](/extension/install/) for connecting it to your account. **Is there a phone app?** Yes — [Jobably Swipe](/app/swipe/) at swipe.jobably.com installs from the browser (Safari: Share → Add to Home Screen; Chrome: the Install banner). It deals your matches as cards: right to keep, left to pass. **Why did a posting score what it did?** Click the score anywhere it appears. The panel shows the written rationale, what the scorer found on seniority, location and pay, and links to the settings that produced it — with a re-score button. **Where do I set the minimum score I see and get emailed about?** Settings → Notifications. The **Matches score floor** is yours for every search at once; the **email alert level** decides which of those are emailed. Each search also has its own threshold — see [thresholds](/scoring/threshold/) for how the three relate. **More questions:** [jobably.com/faq](https://jobably.com/faq) or the [contact form](https://jobably.com/contact). ## What's new https://docs.jobably.com/changelog/ > Recent changes to Jobably. ## November 3, 2026 - **Your words in every section.** The rule that keeps the posting’s word out of a line your record never uses now covers every section of a draft and every studio proposal — the summary, competencies and skills sections included, not only experience. A team-size line the summary was asked for counts as your own words. - **Competency rows are reordered, never rewritten.** A competency row in a tailored draft keeps every one of your terms, in an order that leads with what the posting asks for. A studio proposal that adds, renames or drops a term, or carries a note instead of a term, is held back with the reason. - **A line appears once, and the summary keeps its length.** A bank entry can no longer be added to two sections of the same draft. The summary admits one added paragraph at most — the team-size line when the posting asks for one — and when every paragraph of your summary opens with a label, an addition is asked to as well. - **An opener is never softened.** A rewording that turns “Drove” into “Contributed to” is reverted on a draft and held back in the studio; the record’s opener stays, and the note names it. - **What the posting opens with leads your current role.** When a posting’s first sentence about the role names a duty and a bullet in your current role answers it, that bullet now sits in the role’s top three, and the draft says it moved. Team-size counts, and the bullet budget, read short acronyms (KPI, SLA) and plurals when ranking what to keep. ## November 2, 2026 - **The studio keeps your words and your length.** The tailoring studio now runs the same two rules a one-shot draft does. A proposal that swaps in the posting’s word for one your line never uses, or appends the posting’s own sentence to yours, is held back with the reason — you can still write it yourself. And the bullet budget applies when a session opens: each bullet left out is marked “left out for length” so you can put it back, and the built document reports what was left out. - **Room for a stronger line.** When a tailored draft brings in a bullet from your accomplishment bank that answers the posting, the budget now makes room for it by leaving out the weakest bullet in that role, named beside the draft, instead of keeping the role untouched. - **Team size, stated when it is asked for.** When a posting asks for leading people and your record states your team size, the summary now states it, with the figures exactly as your record has them — on a one-shot draft and in the studio, where it is proposed from your bank for you to accept. ## October 30, 2026 - **Start from your résumé, in one step.** Upload your résumé and Jobably proposes your profile and a first search from it — your current titles, the level they sit at, and where your current role is based — saying where it read each one. Add a salary floor and how you like to work if you want, then choose Start my search: your first scan runs at once and your first matches arrive within minutes, scored. Your AI agent can do the same from the conversation: “Set me up on Jobably — here’s my résumé.” It shows you the proposal, asks what a résumé cannot say, and saves it when you say yes. Agents can now start scoring your new matches too. - **Bring your own agent: ChatGPT, Claude and any MCP client.** Connect the AI agent you already use. In ChatGPT, claude.ai or Claude Desktop, add https://mcp.jobably.com/mcp as a connector and sign in — Jobably shows you who is asking and what it may do, and you can allow it to read only. Codex, Copilot, Cursor and Claude Code connect with one command from Settings → Integrations. Disconnect any agent there at any time. - **A short list every morning.** Each day Jobably picks the best new matches that clear your bar and holds them as the day’s short list — the same list in the app (Short list, in the navigation) and in your agent. Each job shows its score and the reason for it, where it stands, and any documents made for it; a good match you have not acted on carries over to the next day. - **Reports and lists your agent builds for you.** Ask your agent for a weekly pipeline review or a list of jobs to apply to this week, and it saves it to your account. Reports stay live — every number is read fresh when you open one under Reports. Your agent can also show you a tailored resume as text, record what you applied to, and set up a new account from the conversation. - **Getting started, smoother.** Drop your résumé onto the upload area (or choose it) and see which file is ready. Notes from reading your résumé fold into one line you can open if you want the detail. The AI interviewer now follows the conversation as it replies, and when it shows its recap you can still tell it what to change — or open every answer in the form and edit it there. - **Everything on the record.** Settings → Integrations → Activity lists every change to your account — by you, a token or a connected agent — with who did it and how it turned out. A change that cannot be recorded is not made. - **One job, one card — even under two company names.** An employer often posts the same role under two names on two boards — a parent and a subsidiary (Merck and MSD), or a regional arm (Ricoh and Ricoh USA). Jobably now recognizes the repost when the title and the location match too, and shows the role once. Employers on ADP career sites now show their real name instead of an ADP reference number, read from the posting itself. - **New matches are scored first, and faster.** When many matches are waiting for a score, Jobably now scores the newest first — the ones at the top of Matches and on your short list — and works through several at a time. Asking to score again while a run is going simply shows that run. - **Tailoring keeps your current title on your current role.** When one employer lists several titles above shared bullets — a promotion ladder — the tailoring studio now treats those bullets as belonging to the whole stint, so your most recent title is never presented as an earlier, junior one. - **A link to a job lands on that job.** A link from your agent, the short list or the search box opens Matches with that job first, expanded and scrolled into view — however long the list and whatever it is filtered to. Dismiss or pursue it and the next job takes its place. A job you already passed on offers Restore to Matches, and Close to put it away. ## September 15, 2026 - **The Dashboard reads the search as a job.** Three new sections. Where they go: the funnel from postings seen through scored, matched at your floor, drafted, applied, screen or beyond, and closed — each step as a share of the one above. Hit rate by search: the share of each search’s postings that reached your floor, the free way to tell a thin market from a weak query. What it has cost this cycle: credits spent, per match found and per application sent, from the same reading your plan’s meters use. - **Resumes: every document by application, yours beside ours.** The Resumes page now groups every document by application — the tailored resume, cover letter and prep packet Jobably wrote (ours) beside the final you actually sent (yours: upload it on the group under your own filename). Diff a draft against your master, or against your final to see what you changed by hand. Revise reopens the studio with your decisions kept; Re-draft from scratch discards them and queues a fresh draft. A final you actually sent can be marked an exemplar, and later drafts learn its voice and structure — a draft nobody sent cannot. - **A Screen stage, and what came back is its own axis.** The Board gains a Screen column between Applied and Interview for the recruiter screen or phone call (optional — straight to Interview is fine). From Applied on, every card has an “ended:” row — rejected, withdrawn, accepted — because their decision is a different fact from how far you got. An ended application leaves the columns for a Closed list under the board that says both: rejected · reached interview. The Dashboard’s new “What came back” tiles tally still open, rejected, withdrawn, accepted, the average days to hear back and the share answered at all. - **Pursuing: everything between "I want this" and "I sent it", in one place.** A new Pursuing screen, between Matches and Resumes in the navigation. Each posting you are pursuing shows its stage — Queued → Drafted → Reviewed → Final → Applied — worked out from what actually exists, and one button that changes with the stage: Tailor resume, Review & finalize, Decide & build, Mark applied. Drop returns it to Matches. Expand a row for the reasoning behind the score and what the posting is buying, the full description, and your documents. The Board’s Pursuing and Tailored columns now mirror this screen instead of carrying their own controls. - **Profile moves onto the left navigation, with Master Resume and Your Bio under it.** Profile is now the first section in the left navigation — folding open to Master Resume, Targeting, Interview, Exclusions, Accomplishments and a new page, Your Bio — instead of hiding behind the avatar menu. Your master resume lives under Profile (it is who you are); Resumes keeps your tailored drafts. Your Bio is a few sentences in your own voice that your cover letters speak in and that doubles as a draft for your LinkedIn About; “Propose a draft from my record” writes one from your resume and interview answers, and saving runs the same no-fabrication check as a tailored line — a number that is not in your record is refused by name. - **Tailoring now leads with what the posting is buying.** Every score now reads the posting’s orientation from its responsibilities — is the employer buying a leader (own, define, build the team), a doer (execute, coordinate, maintain), or an analyst (model, forecast, report)? A Director title on a doer description is a doer posting. The why-this-score panel shows the reading; it never moves the score. Tailoring and the studio order the bullets within each role to lead with it — order and emphasis only, never a changed claim — and every draft says what the posting reads as, what the draft reads as, and flags a mismatch for you to consider. - **An application that has gone quiet shows its age, not a nudge.** Five days after you mark a posting applied with nothing back, its Board card and the Dashboard now say how long it has waited instead of asking you to follow up — on most application portals there is nothing to send, and the count is for knowing when to write one off. The Dashboard section is now “Waiting on a reply”. - **Matches opens by stating what the scan is looking for.** A panel above your matches says, in plain words, what your searches fetch — the roles and seniority, where, the employer size floor, and the score a match must clear to be shown — and expands to each search’s title phrases, description filter, how many postings it has found and when it last ran, plus the employers and titles you never see. It is read from your saved searches, so it always describes what actually runs. ## September 4, 2026 - **One tailored resume is one tailoring credit.** Tailoring a long resume takes many model calls, and each call was counting as a credit — so a plan’s monthly tailoring allowance ran out after a handful of resumes. A tailored resume now costs exactly one credit: the one-shot tailor, or a studio session including its build and any revision. Your usage history is unchanged, and your credit count has been reset so the earlier over-counting does not carry into this cycle. ## September 3, 2026 - **The tailoring studio: tailor your resume with Jobably beside you.** Tailor resume now opens a studio instead of running in the background. Jobably reads the posting, then walks your resume line by line and role by role, proposing rewordings and additions from your accomplishment bank with the reason each one serves this posting. You accept, reword or keep every line; any line can be edited or left out of the draft. On your own connected model you also get structure and layout recommendations and a headline aimed at the role. Every proposal is checked against your record before you see it, and anything held back says why. Build the document when you are done: your own Word file, a matching cover letter, and the same read-before-you-send step. Reach it from Resumes → Tailor for a job in the left navigation, or from any match or board card. - **Jobably Swipe: review your matches from your phone.** A small installable app at swipe.jobably.com, for one job: every posting your searches found and you have not judged yet, dealt as a card that reads like a profile — company, role, pay, the score and why. Swipe right to keep it under Matched on your board (even one that scored below your floor), left to pass with a reason that tunes future scores; undo is one tap. The Searches tab lets you change threshold, cadence, location and pause from your phone. Signs in with your Jobably account, including Google and Microsoft. Everything else stays in the full app. - **Installing Swipe is one tap, and the deck shows you the gesture.** Where your browser offers to install, a banner at the top of Jobably Swipe adds it to your home screen in one tap; the Account tab has the same button, and the exact steps for your phone when the browser offers none (Safari: Share → Add to Home Screen). The ✕ and ♥ buttons now stay on screen whatever the card’s length, the pass-reason picker slides over the card instead of replacing them, and the first card of a session nudges right and left with its Keep and Pass stamps so the swipe shows itself. Swipe also has its own place on jobably.com — first in the menu. - **Every score explains itself.** Click any score — on Matches, the Board, a card’s drawer, or the Dashboard — and a panel shows the written rationale, what the scorer found on seniority, location and pay, and where to change the things that produced it: your interview answers, targeting and exclusion lists, a search’s threshold, or your Matches floor. Re-score the posting from the same panel once you have. - **Your searches now look like they are searching.** The Dashboard and the Searches page show the work as it happens: what is scanning right now, when the last sweep ran, when the next one is due, how many postings today’s sweeps checked and how many were new — and a feed of the recent sweeps themselves. Every search carries its own “last scan / next in” line, and the page keeps itself current without a reload. - **Search and notification settings are yours to set.** Settings → Notifications now holds a Matches score floor of your own (one number for every search, or leave each search’s threshold in charge), the email alert level, how often match emails arrive (as they are scored, once a day, or off), and how often the pipeline summary is sent (daily, weekly, or off). - **Dismissed postings leave the list.** On Matches, a posting you dismissed while the status filter was on “All” stayed in the list until the filter changed, which read as the dismissal not taking. “All” now means every live status; dismissed postings have their own filter. The card drawer on the Board also names each status change correctly in its history. - **Connected assistants can run the studio too.** Seven MCP tools open a tailoring session, run the review, read every proposal with its reason and guard verdict, save your decisions (accept, reword, keep, leave out, headline, section order), build the document, or start over — the same session the app shows, so you can switch between Claude and the page at any point. tailor_resume still queues the one-shot background draft. - **Numbers on a line can no longer drift.** In the studio, a reworded line must carry exactly the numbers it had — a headcount, a team size, a count of suppliers — not only the dollar amounts and percentages the provenance check already protected. A proposal or an edit that changes one is refused with the reason. ## September 2, 2026 - **A section kept in your own words now says why.** When a tailored draft leaves a section exactly as you wrote it, the reason is shown beside its name: there was nothing to tailor, the time budget ran out before that section’s turn, or every rewrite failed the no-fabrication guards. A section skipped on purpose and one that ran out of time are different facts, and they are no longer shown as the same one. - **Long sections on your own model no longer come back untailored.** On a connected frontier model, a dense section could hit the output limit mid-sentence, and every retry hit it the same way — the draft then kept your original wording and said only that the reply was invalid. The limit now scales with the model and the section, a cut-off reply is retried with more room, and the reason beside the section says it was cut off rather than blaming the reply. If nothing at all could be tailored, the run fails with a try-again instead of handing your own resume back as a draft. - **Review items can carry a caution, and the revision is held to it.** When a suggested phrase can go in only so far, the review panel names what must not be claimed. Approving the item applies that limit to the revision as well: the rewrite is bound by the caution, not just informed by it. A revision that drops a line you already approved now restores it and tells you. ## August 30, 2026 - **Review a draft before you send it: approve, decline, or reword each change.** Beside every tailored resume, Jobably now lists what the posting asks for that the draft doesn’t say — sorted by whether your own record can support it, with the evidence quoted. Approve the items you stand behind (or supply your own wording), and only those lines change; everything you didn’t touch comes back exactly as it was. Changes that can’t be made truthfully are declined with the reason, and any edit you didn’t ask for is reverted automatically. Available on Matches and the Board, and to connected assistants as revise_tailored. - **Drafts tell you everything they did — and did not — do.** A tailored resume now shows which sections kept your original wording, advisory style notes (observations, never silent corrections), and its mechanical scorecard summary — so a partly-tailored draft can never pass for a fully tailored one. - **A tailored resume tells you which layout it was rendered with.** Drafts are normally rendered into your own Word master so your fonts and layout carry through. When that is not possible — the master could not be read, or was not a Word file — the draft is built from a clean generic layout instead, and the Matches and Board pages now say so beside the download, with the reason. Before, both cases looked identical. - **Every change to a search is recorded — including who paused it.** Creating, editing, pausing, resuming or archiving a search now leaves a record of what changed, whether you did it from the app or through a connected assistant, and when — and Jobably’s own automatic pauses are recorded alongside. Ask for a search’s history at GET /v1/searches/{id}/audit. Resuming a search also clears its old automatic-pause note. - **Score threshold and email alert level are named apart.** A search’s score threshold decides what appears on Matches; the email alert level in Settings decides which of those you are emailed about. Both places now say which number they are and point to the other, so a change to one is never mistaken for the other. ## August 27, 2026 - **Matches below your score threshold are counted, never silently hidden.** When your search’s score threshold keeps matches off the Matches page, a strip now says exactly how many — "N matches scored below your search’s score threshold" — with a Show them link that reveals every one in place, badged "below threshold" so you can tell them apart. Nothing was ever deleted; now nothing is invisible either. The Board’s Matched column shows the same count and points to Matches, and connected assistants get the count on every list_matches call plus an includeBelowThreshold option to read the hidden ones. ## August 12, 2026 - **The accomplishment bank is now a pill editor.** Profile → Accomplishments shows every entry as its own pill, so you can see exactly where one accomplishment ends and the next begins. Press Enter to add one, paste a whole list to add one entry per line, and remove any entry with a click. Commas never split an entry — "Negotiated $4.8M savings, 20% on $24M of spend" stays one accomplishment, exactly as tailoring will use it. Each entry is stored individually, and the count on screen is the count that is saved. - **Tailoring handles competency sections that span multiple lines.** A Core Competencies block that wraps across several lines — the way most resumes format it — now reorders cleanly toward the posting. Every term keeps your own spelling, terms never merge across line breaks, and the no-fabrication guard reads the list term by term. - **Adding accomplishments from an AI assistant is more forgiving.** The add_accomplishments MCP tool now accepts the bank as a proper list, as a pasted block with one accomplishment per line, or as a JSON-encoded list — whichever shape your assistant sends. Entries are never split on commas, duplicates are skipped, and unusable input gets a clear error instead of a silent guess. The tool reference shows worked examples for every argument shape. ## August 11, 2026 - **Import your career history from LinkedIn.** Profile → Accomplishment bank now takes your LinkedIn data export. Ask LinkedIn for your archive under Settings → Data privacy → Get a copy of your data, choose the full archive, and upload the zip they email you. Every role description you wrote comes in as accomplishments the tailor can draw on — usually far more than fits on a resume, which is what lets a tailored draft reach for the right result instead of the one that happened to be on the page. Nothing is replaced: entries are added, duplicates skipped, and the count is reported. - **Tailored resumes carry your name and contact details.** Your name, email and phone now come straight from your master resume and appear at the top of every tailored draft, exactly as you wrote them. - **Tailored resumes tell you what they could not cover.** Beside a finished draft you now get the list of things the posting asks for that your record does not evidence. The draft never claims them, and knowing them before you apply is the point — it is also the fastest guide to what is worth adding to your accomplishment bank. - **Core competencies stay true to your record.** A tailored resume now only lists competencies that appear in your own master resume, ordered by what the posting cares about. A skill the record cannot back never appears, so nothing on the page is a claim you would have to defend cold in an interview. - **More of the engine is tunable without waiting for a release.** The models behind resume parsing, saved-posting capture and interview prep, along with the prep research budget and tailoring retry limits, are now operator settings. ## August 9, 2026 - **New searches get time to find their footing.** A new search now has a couple of days to find its first matches before it can pause itself, and a search that has already found matches keeps running through quiet periods rather than pausing at all. - **Searches that had been inactive resume reliably.** A search you had paused for a while now picks up where it left off on its next run. - **Filters behave the same everywhere.** Every filter is available in the app, over the API and through a connected assistant, and behaves identically in each. ## August 8, 2026 - **Every search filter now has a control.** Under "More filters" when you create or edit a search: years of experience, work type, company size, industries, companies to include or skip, and title keywords to skip. Years of experience is the one worth setting — it is the level filter most company career sites use, and those are where most of your matches come from. Job-board levels are individual checkboxes now, so you can pick exactly the levels you want. - **Search for contract and part-time work.** Searches are no longer limited to full-time roles. You can also choose whether to include staffing agencies, and set a maximum company size as well as a minimum. - **A comma between search terms means "or".** Write your target titles as a list — procurement lead, sourcing director, vendor manager — and the search looks for any of them. The query helper shows you how your query was read, and flags a stray word beside an "or" list, which is usually not what you meant. Capitals make an operator: type OR, AND or NOT to combine terms. - **The simple editor leaves complex queries alone.** A query with brackets now opens in the advanced editor rather than being simplified into chips, so what you wrote is what runs. - **Clearer answers when a search cannot work.** Description filters are checked when you save, and you can see how yours will be read. If a search cannot return a result, you are told why — during setup too, which also now uses the locations you gave rather than assuming a country. - **Results say when there are more to see.** A one-off search that hits its limit says so, with a total where we can get one, and you can choose which supply to search: company career sites, job boards, or both. Connected assistants can use every filter, and your exclusions apply automatically. - **Editing a search keeps all of its filters.** Filters that have no control on the page — including any set for you by an assistant — survive an edit untouched. ## August 7, 2026 - **Board and Matches are one workflow.** Tailor a resume straight from any board card, move cards back to Matched from Pursuing or Tailored (before you apply, every move is reversible), and sort columns by score or recency. - **The dashboard navigates.** Score-distribution bars link to Matches at that score and up, pipeline chips open the board, "new since your last visit" survives a refresh, and scan results say "X new, Y already known". Green / amber / red across search cards and credit meters. - **Archive searches.** Done with a search? Archive it — hidden from the working list, restorable any time with its scan history intact. Searches are still never deleted. - **One-click re-score after profile changes.** Saving your profile offers "Re-score my matches" on the spot — existing scores reflect the old rubric until you do. Example text in form boxes is now clearly marked "e.g.", and the seniority menu is grouped by track. - **Onboard by chatting.** A new AI-interviewer path through onboarding: answer in plain language, one question at a time, skip anything. It covers every topic the form covers and ends in the same place — your rubric in your words and your first search running. Switch to the form any time without losing answers. - **Searches fetch smarter, not wider.** United States by default (blank the Location filter for worldwide), senior-only on job boards unless you say otherwise, your excluded titles dropped before scoring ever pays for them, plus company-size and description filters on the search form. First scans look back a full week. - **Work arrangement on every card.** Remote, Hybrid, or On-site shown right below the job title — with the location beside it when it isn’t remote. Unknown stays blank rather than guessed. - **Search thresholds now truly gate.** A match scoring below its own search’s threshold stays out of the default view, exactly as that search’s settings promise. Jobs you’ve acted on, unscored jobs, and extension saves are never hidden — set the score filter to 0 to see everything. - **The Chrome extension is on the Web Store.** One-click install, automatic updates. Save any job page you open in Chrome to your matches — parsed, deduped, and scored against your rubric. Find it by searching "Jobably" or via the install link in the docs. - **Your intake powers your searches.** The first-search query arrives pre-filled from your title keywords, and the Searches page suggests ready-to-run searches built from your questionnaire — your own words, one click to create, or Customize to adjust anything first. - **Resume-aware onboarding defaults.** Your resume headline pre-fills the seniority floor and target titles — clearly marked as guesses, editable on the spot, and your edits always win. Threshold, cadence, and sources are editable during onboarding too. ## August 6, 2026 - **The onboarding interview.** Ten open questions in your own words — your story, ideal role, non-negotiables, dealbreakers, what your resume undersells, what you never want to do again. Answers go into your scoring rubric word-for-word and stay editable on the Profile page. - **Location scoring, fixed.** A strong-fit role you could not actually take no longer scores high. Wrong-country onsite roles cap at 3, out-of-area at 5, region-locked remote at 4, unstated locations at 5 — still surfaced, ranked honestly, with the gap named in the rationale. - **Compensation floors with teeth.** A posting stating pay below your floor now caps at 5 — a consistent number instead of a vague "scored down". - **Guided search builder.** Boolean syntax is optional everywhere: three plain boxes — match ANY, must ALSO contain, must NOT contain — build the query for you, with an Advanced mode for hand-written expressions. - **Matches: filter by search and by date.** See which of your searches found each match, and narrow to today or the last 3 / 7 / 14 / 30 days — alongside status, score, source, and engagement filters. - **Re-score any match.** A per-match Re-score button with live progress; the fresh score lands on the card the moment it finishes. Finished statuses, including errors, can be dismissed. - **Connect your AI assistant in one paste.** After creating an access token, Settings shows the complete connect command for Claude Code, Codex, or Copilot on Windows, macOS, or Linux — token already filled in. Revoked tokens stay visible, dimmed. - **Feedback that reaches us with context.** The feedback bubble now sends the exact page you were on — filters and all — so reports can be reproduced first try. - **Onboarding and profile, sharper.** Seniority floor covers every function, work arrangement is multi-select, industries to avoid sit beside preferred ones, title keywords you WANT join the rubric, and the resume preview shows the full parsed text. New accounts get a live first-scan card on the dashboard. - **Email, grown up.** Branded, readable email throughout: a welcome note when you join, a what-happens-next note after onboarding, a Monday weekly digest (opt out in Settings), trial reminders, and payment notices. Sign-in codes can be resent from the login screen. ## August 5, 2026 - **Jobably Executive.** A second plan for searches where timing decides: hourly scans, new matches scored automatically as they arrive, roughly five times the credits, up to 25 searches. Switch between plans any time from Billing. - **Application packages.** Tailoring now produces a matching cover letter alongside the tailored resume — both under the same guarantee that nothing is ever invented about you. - **Tailoring that preserves your document.** Tailored resumes are edits of YOUR document, not rewrites: your sections, your dates, your formatting survive, with four independent checks that reject any draft that drifts from the record. - **High-score email alerts.** New matches at or above your threshold (default 8) email you right away — one digest per scoring run, never one email per job. Configurable in Settings. - **Interview prep packets.** Prep now runs on your own Anthropic key with live web research: company brief, likely interview themes, STAR stories from your verified history, and outreach drafts you send yourself. - **Your own model keys, expanded.** xAI and Ollama Cloud join Anthropic, OpenAI, and Google. Keys are verified with the provider before saving, and a Settings picker chooses which model runs your tailoring. - **Pay captured wherever it is stated.** Salary ranges on scanned roles and hourly or project rates on contract gigs — never estimated; blank when the posting is silent. - **Credits and status you can trust.** Live remaining-credit meters on the dashboard and Billing, persistent operation status on every match, and a public status page fed by real nightly checks.