agentleFS
allthriveaiallthriveai/amplify/.claude/skills/wiki/SKILL.md

wiki

Skill0 starsChanged 3 months agoMITView on GitHub

Answers a question from the wiki, with citations. Reads the index first, drills into the pages it names, synthesizes an answer, and offers to file anything worth keeping. Use when the user runs /wiki, asks what they know about something, or asks a question their vault should be able to answer.

42 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ai-slop-cleaner, ralplan and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Jurel89Skill

    ai-slop-cleaner

    Clean AI-generated code slop with a regression-safe, deletion-first workflow and optional reviewer-only mode

    Also in 4 projects that use this

  • ibahgatSkill

    ralplan

    Create a strict, quality-first execution plan with explicit gates and rollback-ready sequencing.

    Also in 4 projects that use this

  • Yeachan-HeoSkill

    ultragoal

    Create and execute durable repo-native multi-goal plans over Codex goal mode artifacts.

    Also in 3 projects that use this

  • woosikkimSkill

    deepinit

    Generate hierarchical AGENTS.md codebase maps for agent context

    Also in 3 projects that use this

  • ibahgatSkill

    hud

    Control omi visual HUD profile and keep status output readable in normal, compact, or hidden mode.

    Also in 3 projects that use this

  • klauswgSkill

    deep-interview

    在明确批准执行之前,通过数学化歧义门控进行苏格拉底式深度访谈

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 36 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read .amplifyrc for paths.wiki and paths.sources.

    line 14
  2. Very specific:

    Check the updated dates. A page that has not been touched since a source that supersedes it is worth flagging in the answer.

    line 30
  3. Very specific:

    If the wiki cannot answer, read the clippings in {sources}/.

    line 35
  4. Distinctive:

    If this keeps happening on a topic, the wiki has a hole worth filling deliberately.

    line 38
  5. Distinctive:

    When pages disagree, say so and give both, rather than silently picking one.

    line 46
  6. Distinctive:

    No hedging filler. If the answer is uncertain, name the specific uncertainty.

    line 48
  7. Very specific:

    If the answer produced a real comparison, a resolution, or a connection nobody had written down, it is a synthesis page.

    line 52

“Common” is measured: kept by at least 6 in 10 of the 36 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

Query the Wiki

The wiki exists so knowledge is compiled once and kept current, not re-derived from raw sources on every question. Read it in that order: index, then pages, then raw material only if the wiki genuinely cannot answer.

Step 0: Load configuration

Read .amplifyrc for paths.wiki and paths.sources. Read {vaultPath}/CLAUDE.md for the conventions.

Step 1: Read the index first

{wiki}/index.md is the catalog. Read it before opening anything else. It exists so you can find the three relevant pages without scanning two hundred files.

Pick the pages that plausibly bear on the question — including ones that might contradict the obvious answer.

Step 2: Read those pages

Open them. Follow [[links]] one hop when a page points at something clearly relevant. Do not crawl the whole graph.

Check the updated dates. A page that has not been touched since a source that supersedes it is worth flagging in the answer.

Step 3: Go to raw sources only if you must

If the wiki cannot answer, read the clippings in {sources}/. Then say so plainly: that is a gap, and the fix is a wiki page, not a longer answer. Offer to write it.

If this keeps happening on a topic, the wiki has a hole worth filling deliberately.

Step 4: Answer with citations

Cite wiki pages as [[Page Title]], inline, where the claim is made. The user should be able to click straight to the source of any sentence.

  • Say what the wiki actually supports, and where it is thin.
  • When pages disagree, say so and give both, rather than silently picking one.
  • Separate what the sources claim from your own inference. Label the inference.
  • No hedging filler. If the answer is uncertain, name the specific uncertainty.

Step 5: Offer to file what is worth keeping

If the answer produced a real comparison, a resolution, or a connection nobody had written down, it is a synthesis page. Offer to save it to {wiki}/Synthesis/:

---
tags: [relevant topics]
sources: [clipping filenames behind it]
created: [today]
updated: [today]
---

# [Title Case Question or Comparison]

[The question being resolved.]

## Where the evidence lands

[What each side holds, cited to [[pages]], and where it comes out.]

## Still open

[What would change the answer.]

Then add a line to {wiki}/index.md and append to {wiki}/log.md:

## [YYYY-MM-DD] query | [The question]
Answered from [[Page A]], [[Page B]]. Filed [[New Synthesis Page]].

Log the query even when nothing gets filed — a run of queries the wiki could not answer is the clearest signal of what to ingest next.

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.