agentleFS
MaximSTHMaximSTH/meta-layer-starter/.agents/skills/doc-consistency/SKILL.md

doc-consistency

Skill2 starsChanged 5 months agoMITView on GitHub

Walk the doc-consistency sweep when a session makes a substantive edit to a draft or active planning/spec doc in markdowns/ (decision changed, shared value updated, section restructured), or on explicit /doc-consistency invocation. Not for typo fixes, status flips, or style/voice issues — those belong to mechanical gates.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

/doc-consistency

Substance lives in markdowns/protocols/doc-consistency.md.

Steps:

  1. Discriminate kickoff vs iteration per auto-trigger-discriminator.md. Typo fixes, status flips, and style edits exit silently. Explicit /doc-consistency invocation bypasses.
  2. State the semantic delta of the edit — what claim changed, what premise it may invalidate.
  3. Enumerate the sweep set: all status: draft + status: active docs; the edited doc itself, in full (sibling sections + summary surfaces — frontmatter description, intro, verdict/summary tables); every declared quoter (quotes: frontmatter or <!-- quotes: ... --> comment) of any changed value.
  4. Read each enumerated doc and check it for dependent claims — restated values, descriptions of the changed mechanism, premises the change invalidates.
  5. Classify findings — contradiction / stale-but-harmless / unaffected. Fix contradictions in the same change-set; report anchored file:line per rubric-shared-anchors.md.
  6. Check the handoff + escalation triggers. If a value moved doc → runtime, verify the value-authority handoff edit landed. If the same doc pair contradicts repeatedly or the sweep set has grown heavy, walk the protocol's escalation path (consolidate first; gate only if consolidation doesn't apply).

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.