agentleFS
TJaenichenTJaenichen/claude-code-harness/skills/user/q/SKILL.md

q

Skill0 starsChanged 3 months agoMITView on GitHub

Answer a question and nothing else — no context, no caveats, no call to action, no footer. Use when the user invokes /q or prefixes a prompt with "q:".

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No preamble: Don't restate the question, don't set the scene, don't say what you're about to do.

    line 18
  2. Distinctive:

    If the question genuinely cannot be answered without a real investigation, say so in one sentence and stop.

    line 31

≈ “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

Answer-Only Mode

The user wants the answer, and nothing else. They are not asking you to investigate, implement, review, or plan — even if the question touches code that obviously needs work.

Answer this: $ARGUMENTS

Rules

  • No preamble. Don't restate the question, don't set the scene, don't say what you're about to do.
  • No surrounding context. Skip background, history, and related-but-unasked detail. Include a caveat only when the bare answer would be actively wrong without it — then make it one clause, not a paragraph.
  • No call to action. No "want me to fix that?", no suggested next steps, no offering to open a PR or write a test.
  • No end-of-turn orientation footer.
  • Minimal tool use. Answer from context if you can. If the answer is genuinely unknowable without looking, make the fewest calls that settle it, then give just the answer — not a tour of what you found.
  • Length: one or two sentences. A single word, number, file path, or file.cs:42 reference is a complete and ideal answer.
  • If the question genuinely cannot be answered without a real investigation, say so in one sentence and stop. Don't start the investigation.

Formatting

Plain prose or a bare value. No headers. A short list only if the answer is inherently a list (e.g. "which three configs?"). No bold-label summary lines.

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.