agentleFS
Sign inStart free
Surething-ioSurething-io/cockpit/skills/ss/SKILL.md

ss

Skill44 starsChanged 8 months agoMITView on GitHub

Find a past Cockpit session — any project, any engine, any date — from a one-line description, and link to it.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Write 3–8 short keywords that would literally appear in that conversation:

    line 11
  2. Distinctive:

    Read the snippets; ranking is keyword-based, the judgment is yours

    line 39
  3. Distinctive:

    No hits → try synonyms, the other language, or shorter keywords

    line 41
  4. Distinctive:

    Stop after 3 rounds; say what you tried

    line 42
  5. Very specific:

    If link is null, give the cwd and sessionId instead.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Find the past session the user is describing. The user rarely remembers the exact words, so do NOT search their sentence verbatim — expand it into keywords, search, read the snippets, and judge.

1. Expand the description into keywords

Write 3–8 short keywords that would literally appear in that conversation:

  • The distinctive nouns of the topic (a file, API, error, product, library name)
  • Both languages when the topic is technical: e.g. 快照 and snapshot, 跨站 and CSRF
  • Synonyms and the words the assistant would have used in its reply
  • Short is fine — 2-character Chinese words match. Avoid generic words (问题, 代码, fix) that hit hundreds of sessions

2. Search

Each q is one keyword; repeat q for more (or join alternatives with |). Always URL-encode:

curl -fsS --get "{{BASE_URL}}/api/sessions/search" \
  --data-urlencode "q=<keyword 1>" \
  --data-urlencode "q=<keyword 2>" \
  --data-urlencode "q=<keyword 3>"

Optional filters:

  • --data-urlencode "cwd=/abs/project/dir" — only that project (and its subdirectories)
  • --data-urlencode "since=2026-09-01" — only sessions updated since (ISO date or epoch ms)
  • --data-urlencode "limit=20" — default 10, max 30

Response: results[], best first, each with engine, cwd, title, updatedAt (epoch ms), matchedTerms, matchCount, snippets, and link (null when the project directory is unknown). corpus reports indexing stats; the first call on a machine may take several seconds while the corpus is built.

3. Judge, refine if needed

  • Read the snippets; ranking is keyword-based, the judgment is yours
  • Too many plausible hits → add a distinctive keyword, or use cwd / since if the user hinted at a project or time
  • No hits → try synonyms, the other language, or shorter keywords
  • Stop after 3 rounds; say what you tried

4. Reply

Give 1–3 candidates, most likely first. For each: title, project (basename of cwd), engine, date, one line on why it matches, and the link exactly as returned, as a markdown link:

1. **<title>** — <project> · <engine> · <YYYY-MM-DD>
   <why this is the one>
   [Open session](<link>)

If link is null, give the cwd and sessionId instead. Never invent or rewrite a link.

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.

Check your setup, free