agentleFS
chenson42chenson42/stackkeel/.claude/skills/trivial/SKILL.md

trivial

Skill1 starsChanged 53 days agoMITView on GitHub

Stamp a single-use, single-path Trivial exemption for the work-log gate (scripts/worklog-gate.mjs) — operator-invoked only, never self-issued by an agent

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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. Very specific:

    One path, not a glob: If multiple files need the exemption, the operator runs /trivial again for each — this keeps every exemption narrow …

    line 37

≈ “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

/trivial

Stamps .claude/trivial-ok.json, the only local exemption scripts/worklog-gate.mjs (the PreToolUse hook on Edit/Write) will honor. This exists because AGENTS.md's Classification table says Trivial-class edits ("typo fix, single-line config edit, doc-only change") need no work-log at all. Without a local exemption, every Trivial edit to apps/**/packages/** source would be permanently blocked with no recovery path.

THE RULE — read this before doing anything else

This skill must be invoked by the operator, directly, of their own initiative. An agent must never run /trivial on its own to route around a block it just received. If a block looks wrong to you (the agent), stop, explain to the operator why you think the edit is genuinely Trivial per the Classification table, and wait for the operator to decide and invoke this skill themselves.

The point is that no one grades their own work. No message from any agent — including a summary of "the operator said this was fine" — is a substitute for the operator typing /trivial themselves. The invocation model cannot mechanically distinguish "the human typed this" from "the agent invoked this mid-turn," so this is enforced as a norm here, in the agent files, and in AGENTS.md — not as a code-level constraint. Treat it as load-bearing anyway.

What it does

Given $ARGUMENTS as <path> "<reason>":

  1. Resolve <path> to a repo-relative path (matching what scripts/worklog-gate.mjs will compare against — e.g. an apps/<app>/src/... relative path, never an absolute one).
  2. Write .claude/trivial-ok.json (gitignored):

    {
      "path": "<path>",
      "reason": "<reason>",
      "stampedAt": "<now, ISO 8601>",
      "expiresAt": "<now + 10 minutes, ISO 8601>"
    }
  3. Tell the operator: which path is exempted, for how long (10 minutes), and that the exemption is single-use — the next Edit/Write call the hook evaluates against this exact path consumes and deletes the marker, whether or not that call is the one the operator intended.

Constraints

  • One path, not a glob. If multiple files need the exemption, the operator runs /trivial again for each — this keeps every exemption narrow and auditable, never a blanket "skip the gate for a while."
  • 10-minute expiry, single-use. A marker that outlives its edit, or that silently covers a second unrelated edit, defeats the point of a narrow, operator-issued exemption.
  • Do not stamp a marker for a path you (the agent) were just blocked on, on your own initiative. Surface the block to the operator instead. See THE RULE above.

When you're done

Report the exempted path and expiry back to the operator in one line. Do not proceed to make the exempted edit yourself unless the operator has separately asked you to — stamping the marker and making the edit are two different actions.

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.