agentleFS
Miracle-AlignerMiracle-Aligner/job-search-hq/.claude/skills/recheck-statuses/SKILL.md

recheck-statuses

Skill1 starsChanged 5 months agoMITView on GitHub

Use when the user wants to verify which Job Search HQ vacancies are still live vs closed / deleted. Triggers: /recheck-statuses, "check dead listings", "clean closed roles", "are these still open?", "refresh statuses". Walks `data/vacancies.json`, opens each URL (web_fetch first, Chrome fallback for auth-gated), and flips dead postings to `status: closed` — without ever overwriting a real pipeline status (`applied`, `hr_screen`, `tech_interview`, `final`, `offer`, `rejected`).

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use today's ISO date.

    line 63
  2. Very specific:

    Never touch score, ss, plus, minus, summary, cv, note, altLinks.

    line 66

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “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

recheck-statuses

Flow

  1. Segment entries into sweepable vs pipeline-locked.
  2. Probe each sweepable URL (cheapest tool first).
  3. Apply closed to dead postings, preserving prior statusDetail.
  4. Report + python3 scripts/validate.py → python3 scripts/bake.py.

Segmentation

Statuses are about the candidate's pipeline, not the posting's liveness — they don't undo each other. Only flip rows that are currently:

  • to_apply — not yet engaged
  • watch — already on hold
  • closed — already dead (re-verify; postings sometimes reopen)

Skip rows in any of these stages, even if their URL looks dead:

  • applied, hr_screen, tech_interview, final, offer, rejected

The reasoning: a posting going dark doesn't undo a real conversation. If the user wants to reopen one of these, they'll say so explicitly.

Probing

Primary URL hostTool"Closed" signal
jobs.dou.ua/companies/...WebFetchHTTP 404 · вакансія неактивна / закрита · page redirects to /companies/<co>/vacancies/
djinni.co/jobs/<id>-<slug>/WebFetch, Chrome fallback404 · Вакансія неактуальна · архівована
linkedin.com/jobs/view/...Claude in Chrome (auth)"No longer accepting applications" · 404; if not logged in, mark unknown, don't guess
Google Docs / OtherWebFetchTitle + role text present → alive; else unknown, ask the user

Throttling: DOU 429s under parallel load. Cap concurrent WebFetchs at ~3. If you hit two 429s in a row, back off 60s and mark the rest unknown instead of looping.

A slug change (same role, new vacancy id) reads as deleted here. If the user has the new URL, they'll re-add via /add-vacancy.

Applying

For each dead row:

{
  "status": "closed",
  "statusDetail": "<b>Auto-closed YYYY-MM-DD by /recheck-statuses</b> — <reason>. Was: <prior statusDetail>."
}
  • Use today's ISO date.
  • Quote the reason specifically: HTTP 404, вакансія неактивна banner, redirected to company list, URL repurposed for different role, etc.
  • Preserve the prior statusDetail text inside the new one — don't lose the original context.
  • Never touch score, ss, plus, minus, summary, cv, note, altLinks. Those are fresh-at-add judgements, not liveness signals.

For rows that came back alive: no change. For rows that came back unknown: no change, but include them in the final report so the user can check manually.

Reporting

Closed (will be auto-marked):
  - <id>  <co>  <role>  →  <reason>
Alive (no change):
  - <id>  ...
Unknown (needs a human):
  - <id>  <co>  <role>  →  <why>
Skipped (active pipeline):
  - <id>  <status>  <co>

Bulk runs operate with pre-approval — apply directly. Single-vacancy runs (e.g. "recheck langate"): apply directly too.

Finish

python3 scripts/validate.py
python3 scripts/bake.py

Pairing

This skill is paired with /refresh-listings (the other direction — find new postings). A common cadence: run /recheck-statuses first to clear out dead rows, then /refresh-listings to top up.

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.