agentleFS
VincenzoImpVincenzoImp/openings/skills/manage-sources/SKILL.md

manage-sources

Skill7 starsChanged 3 days agoMITView on GitHub

Add, tune, pause and review Openings job sources - "add a search for...", "follow company X", "why so much noise from Y?", "which sources work?". Previews every new source before saving and measures sources by what they bring.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are 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.

  • 3

    Very specific

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

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

    Read openings first and call workspace_context: What I'm looking for and What I rule out define a useful job.

    line 11
  2. Very specific:

    Run it once with sources_run when the owner wants results now.

    line 24
  3. Very specific:

    Read why its jobs were dismissed: jobs_list of dismissed jobs, narrowed to its site or company, shows each dismissal reason.

    line 32
  4. Distinctive:

    Prefer the narrowest fix: a more specific query or place, or excluded title words checked against the shortlisted jobs so none would be …

    line 36

≈ “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

Manage job sources

Read openings first and call workspace_context: What I'm looking for and What I rule out define a useful job.

Add a source

  1. collection_get with the source kinds: what each kind needs (a search query and place, a company's board, a feed URL). For a company, find its careers page and use its board kind when one matches; otherwise a search.
  2. Preview it with sources_run: nothing is stored. Read the sample against the Preferences: how many would the owner want? Narrow the query, place or title filters until most are relevant, without excluding jobs they want.
  3. Save it with sources_save, enabled, on a daily schedule unless the owner says otherwise. Add several in one batch and check every item.
  4. Run it once with sources_run when the owner wants results now.

One query and one place per search; split "Zurich or Basel" into two.

Tune a noisy source

  1. sources_list gives each source's counts: found, shortlisted, applied, replied. Judge a source only after it has found enough jobs.
  2. Read why its jobs were dismissed: jobs_list of dismissed jobs, narrowed to its site or company, shows each dismissal reason. A recurring, clear mismatch (another field, a ruled out title or language) is worth a filter; a varied mix is triage's job.
  3. Prefer the narrowest fix: a more specific query or place, or excluded title words checked against the shortlisted jobs so none would be lost. Preview the change before saving.
  4. Pause a source that brings nothing useful; delete one only when the owner asks.

Failures

collection_get lists failing sources and sources_get their recent errors. Provider limits and cooldowns are protective: wait them out rather than raising or lifting them, unless the owner asks. Retry a failed fetch with collection_update once the cause is fixed. Turn collection off only when the owner asks.

Report back

In chat: each source added, changed or paused, with what its preview or counts showed; failures and their causes; what you suggest next, such as a company worth following.

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.