agentleFS
rcdelfinrcdelfin/agentkit/skills/operations/local-wiki/SKILL.md

local-wiki

Skill2 starsChanged 3 months agoMITView on GitHub

Ingest source content (ClickUp cards, meeting notes, transcripts, specs) into the project's governed local knowledge base. Dynamically locates the wiki (prefers docs/wiki/), then obeys the wiki's own AGENTS.md for structure and rules. Trigger: 'ingest into the wiki', 'add to wiki', 'capture this card/note locally', or any workflow that needs to persist a source doc locally before working on it.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Local Wiki

Ingest source content into the project's governed local wiki — the append-only raw corpus + rebuild-only topic pages that capture knowledge before it scrolls away in ClickUp/chat. This skill locates the wiki dynamically and defers to its own governance for the exact structure; it does not hardcode one layout.

Locate the wiki (dynamic — prefer docs/wiki/)

A wiki is "governed" if it has an AGENTS.md (or CLAUDE.md) defining its layout. Search in this order, use the first match, prefer docs/wiki/:

for d in docs/wiki wiki docs/kb docs/knowledge docs/notes; do
  if [ -f "$d/AGENTS.md" ] || [ -f "$d/CLAUDE.md" ]; then echo "WIKI=$d"; break; fi
done
  • Found → set WIKI=<dir>, read <WIKI>/AGENTS.md and obey it (DOX: nearest AGENTS.md wins).
  • Ambiguous (several candidates, none docs/wiki/) → list them and ask the user.
  • None found → ask the user where the wiki lives, or whether to bootstrap a minimal docs/wiki/. Do not silently invent a structure.

Obey the wiki's own governance

The wiki's AGENTS.md is the source of truth for: the raw-corpus layout, page format, what's append-only vs rebuild-only, whether the tree is git-ignored, and URL-sourcing rules. Read it before writing. Never hand-edit rebuild-only pages; never delete from the append-only raw corpus.

Ingestion pattern (adapt to the wiki's actual structure)

  1. Fetch the source via the right skill — clickup for a ClickUp card (show_task.py <ID> + show_comments.py <ID>). For meeting notes / transcripts / specs, take the user-provided content.
  2. Write to the append-only raw corpus at the wiki-defined path, with the wiki-required front-matter. For the standard docs/wiki layout:
    • ClickUp card → <WIKI>/raw/clickup/<ID>-<slug>.md with custom_id: and url: copied from the card (never retype a URL).
    • Other sources → the subdir the wiki's AGENTS.md specifies.
  3. Rebuild the topic page per the wiki's rebuild flow — for docs/wiki: pages/<slug>.md in the canonical meta-block format, then update index.md and append a one-line log.md entry.
  4. Don't touch a consumer-facing changelog (e.g. docs/wiki/CURRENT-CHANGELOG.md) from ingestion — it has its own separate, human-reviewed workflow.

Reference implementation (docs/wiki)

The convention this skill defaults to when the wiki's AGENTS.md is silent:

PathRoleRule
raw/clickup/one file per GYMED-XXX ClickUp taskappend-only
raw/ (other)meeting notes, transcripts, exportsappend-only
pages/one topic page per source, AI-builtrebuild-only — never hand-edit
glossary/domain termsrebuild-only
queries/saved query outputsadd; rewrite only if stale
index.mdcatalogupdate on every page change
log.mdwiki changelogappend-only
CURRENT-CHANGELOG.mdconsumer release notesseparate workflow — don't touch on ingest

Pitfalls

  • Nearest AGENTS.md wins — if <WIKI>/AGENTS.md contradicts this skill, follow the wiki, not this file.
  • Pages are rebuild-only — to change a page, rebuild it from raw/; don't hand-edit.
  • raw/ is append-only — never delete or overwrite an existing export.
  • URLs are copied, never retyped — paste from the source front-matter/body.
  • Often git-ignored / local-only — the whole tree commonly stays on contributor machines; don't assume it's committed.
  • Don't auto-create a wiki — if none is found, ask; bootstrapping a governed wiki is a decision, not a side effect.

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.