agentleFS
MicrosoftKnown publishermicrosoft/amplifier-agent/examples/python/llm-wiki/skills/wiki-maintainer/SKILL.md

wiki-maintainer

Skill9 starsChanged 21 days agoMITView on GitHub

Merge proposals into a concept or entity page, fix lint findings, and write lint reports for the Amplifier ecosystem wiki. Load before writing any wiki page other than a repository page.

3 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

3 rules. About 0 are common, 2 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the current page when it exists.

    line 10
  2. Distinctive:

    Add each proposal where it belongs.

    line 11
  3. Very specific:

    Write the whole page in one write_file call.

    line 18

≈ “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

Wiki maintainer

Merging proposals into a page

  1. Read the current page when it exists. Keep every claim that still has a citation.
  2. Add each proposal where it belongs. Merge proposals that say the same thing into one claim with all their citations.
  3. A contradiction proposal goes into ## Contradictions with the claim it disagrees with. Keep both citations.
  4. A link proposal becomes a link in ## Related or in the sentence it supports. Fill ## Related from the link proposals, related pages in the index, and the pages that link here. Leave it out when nothing is related; never write a placeholder such as "None yet".
  5. Rewrite the summary in the frontmatter when the page now says more.
  6. Write the whole page in one write_file call.

Concept page

# Title

What it is, in two or three sentences.

## How it works
## Where it appears        one bullet per repository page, with what that repository does with it
## Related                 links to other concept and entity pages; left out when there are none
## Contradictions          only when there are any

Entity page

# Name

What it is and which repository publishes it.

## Use
## Interfaces              the configuration keys, commands, or functions a person calls
## Related                 left out when there are none
## Contradictions

Fixing lint findings

Broken link       point it at the right page, or remove the link and keep the text
Missing page      link the closest page that exists, or remove the link
No sources        add citations from the pages this one links to, or from raw/ when you read the file
Stale citation    read the file at the current SHA and update the claim and its citation

Lint report

# Lint report YYYY-MM-DD

## Contradictions
## Missing concept pages
## Broken and missing links
## Stale or missing citations
## Suggested questions

Each finding names the page as a relative link, states the problem in one sentence, and gives one fix. Write "none" under a heading with no findings.

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.