agentleFS
akshay-diwadkarakshay-diwadkar/first-four-hours/skills/sensemaking/scribe/SKILL.md

scribe

Skill1 starsChanged 4 days agoMITView on GitHub

Write source-grounded technical documentation such as runbooks, procedures, setup guides, notices, error messages, and reference pages, or audit an existing document read-only. Use when a document must faithfully carry supplied commands, paths, values, warnings, and recovery without invented facts.

16 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 14 edited versions we read keep these.

  • 1

    Distinctive

    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:

    Write only with document-creation or revision authority.

    line 14

“Common” is measured: kept by at least 6 in 10 of the 14 edited versions of this file we could read. “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

Scribe

Purpose and authority

Turn supplied technical sources into a document its reader can execute and trace. Never invent source facts; preserve bound commands, paths, values, ordering, warnings, and recovery. Validation does not establish independent factual truth.

Write only with document-creation or revision authority; implicit invocation never grants write or remediation authority. Audit is a separate operation that writes only its report. Audit must not modify the document, bundle, or bound sources.

Derive audience, reader job, and scope from the request and sources. Ask the user only for a material publication or content-policy choice the sources cannot settle; a missing source fact is an unresolved gap. Consult reason before committing to a document plan, after a failure or direction change, and before reporting completion. Final and stop reports trace reason checkpoints per its Checkpoint trace.

Write and audit

Each document is a deliverable: in write mode, create or edit the actual file in place where the user wants it. Both operations check every document one document-bundle.json lists against the sources it binds, per Document Bundle and Source Grounding. The bundle is support state: write it from templates/document-bundle.json in a temporary directory outside the repository. List several documents to handle them in one run.

To write, follow Write; its language checks apply MTE-1, an original contract only inspired by ASD-STE100: never claim official ASD-STE100 compliance or imply that this skill contains its approved-word dictionary. The write command only checks the documents and edits no file:

python /absolute/skill-root/scripts/cli.py --repo-root /absolute/repo \
  --input bundle=/absolute/temp/document-bundle.json --format json write

To audit, follow Audit: run audit the same way with --input slug=<audit-slug>. It writes only .first-four-hours/scribes/<slug>/scribe-audit.md, one section per document.

Completion

Write completes only when write returns result.status: complete. On incomplete, only the listed gaps remain: report each with its owner. Otherwise repair the document each diagnostic's document names, or the bundle, as its required_action directs, then rerun. Audit completes only when audit returns status: complete; report the scribe-audit.md path and each document's document_sha256 and findings. A passing result is not completion while a material reader job stays unsupported: report the exact gap and its next owner. End with one optional next step fitting this outcome: a fenced, self-contained prompt naming any repository, issue, PR, or artifact it needs, or say none is needed. Only a "yes" starts it, via any skill it names, but the user runs a charter, diagnose, or ticket step as its /<skill> command.

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.