agentleFS
mukewlmukewl/jobkit/.claude/skills/find-jobs/SKILL.md

find-jobs

Skill1 starsChanged 3 months agoMITView on GitHub

Collect fresh jobs from every configured board and API source, then immediately score and tier them against the user's CV and write them to the tracker. Use when asked to find jobs, run a search, refresh the queue, check for new roles, or see what is out there.

11 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use interview-prep.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AnthropicSkill

    interview-prep

    Create structured interview plans with competency-based questions and scorecards. Trigger with "interview plan for", "interview questions for", "how should we interview", "scorecard for", or when the user is preparing to interview candidates.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 2 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read profile.md and the user's CV at cv/master_cv.* before you start.

    line 11
  2. Very specific:

    Read boards from config.json.

    line 18
  3. Very specific:

    Read the results list. Prefer an accessibility snapshot or document.body.innerText over a screenshot — you want text, not pixels.

    line 21
  4. Very specific:

    Write everything you find to data/browser_jobs.csv with the header:

    line 27
  5. Distinctive:

    Never solve a CAPTCHA and never log in on their behalf: If a board demands either, stop on that board, say which one and why, and suggest …

    line 39
  6. Very specific:

    If boards is empty or all disabled, skip this step silently.

    line 43
  7. Distinctive:

    Do not stop at step 2 and hand over a raw queue.

    line 62

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Find jobs

Collect, then filter. Always both. A pile of unranked listings is not a result — the user asked what is worth applying to, so finish the job before reporting.

Read profile.md and the user's CV at cv/master_cv.* before you start. Work authorisation and language level decide what survives, and the CV decides what scores well.


Step 1 — Boards the user searches themselves

Read boards from config.json. For each entry with "enabled": true:

  1. Open its search_url in a browser tool.
  2. Read the results list. Prefer an accessibility snapshot or document.body.innerText over a screenshot — you want text, not pixels.
  3. Extract per row: title, company, location, url, posted date.
  4. Paginate or scroll if the page has more and it is cheap to do so. Stop after two consecutive scrolls with nothing new, and never more than five pages.

Write everything you find to data/browser_jobs.csv with the header:

title,company,location,url,posted_at

That file is what csv_import reads, so browser results go through exactly the same dedupe and filtering as everything else rather than around it.

When a board fails, say so and move on. Sites break, redirect, or throw a challenge. One dead board must not take down the run — report it in your summary and carry on.

Never solve a CAPTCHA and never log in on their behalf. If a board demands either, stop on that board, say which one and why, and suggest they either log in themselves in that browser first or export results to CSV instead.

If boards is empty or all disabled, skip this step silently. It is optional.

Step 2 — API sources

python search/fetch_jobs.py

This runs the sources in config.json, folds in data/browser_jobs.csv via csv_import, drops anything already in seen_jobs.csv, applies the keyword and location filters, writes data/queue_<date>.json, and rebuilds public/index.html.

Report any [warn] lines rather than hiding them.

Step 3 — Filter against the CV (never skip this)

Now follow .claude/skills/filter-jobs/SKILL.md in full. It is the single source of truth for tiering, and it writes the tracker.

Do not stop at step 2 and hand over a raw queue. The keyword filter in the pipeline is deliberately crude — it matches title words and location strings and knows nothing about whether the user can actually take the job or do it well. The filter step is where the value is.

Step 4 — Report

Tier A roles in full. Notable B ones. Counts for C and X. Which boards worked and which failed. Do not paste the whole queue into the chat — that is what the tracker and the dashboard are for.

Close with the single most useful next action, which is almost always:

/apply-to-job <url of the best tier A role>

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.