agentleFS
smkg75smkg75/wikidoc/skills/wiki/SKILL.md

wiki

Skill2 starsChanged 18 days agoMITView on GitHub

Reads and writes the wiki of the document corpus. Use when a question bears on the user, their entities, dossiers or correspondents, when a document must be found in the corpus, or when a durable fact or a change in a dossier's state surfaces and must be written down.

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

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

  • 0

    Common

    Most of the 36 edited versions we read keep these.

  • 0

    Distinctive

    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. Very specific:

    Read index.md. Each line names the question a file settles.

    line 14

“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

Wiki

$WIKIDOC_HOME/wiki/ (default ~/.wikidoc/wiki/) holds what the corpus cannot say about itself: who is who, which entity was live in which period, which arbitration was made and why. context.md carries the durable facts, decisions.md the dated arbitrations, state.md the present — one current line per dossier in flight, and the open questions — log.md the chronicle of what happened and when, filing-patterns.md and trash-criteria.md the observed destinations and removal criteria. A wiki grows the files its owner needs; index.md names them all.

No wiki/index.md: stop on "Run /wikidoc:setup first".

Read

  1. Read index.md. Each line names the question a file settles.
  2. Open the file that settles this question. The files are long: list the headings, read the section that answers.
  3. Answer with the fact and the place it lives, file § section. Where the wiki is silent, say so, dig in the corpus, and write what the digging established.

Done when the answer names its source, or names the wiki's silence.

Find a document

memory.py find <term> searches what every filed document was said to be: description, tags, identifiers. memory.py show <path|md5> gives the decision behind one file. The script lives in scripts/ at the plugin root (scripts/ sits beside skills/, two levels up from this file). On macOS Spotlight carries the same descriptions and tags, and also covers what no pass has judged yet.

Done when the path handed to the user has been re-stat'ed on disk.

Write

The wiki is written in any session, the moment the fact surfaces: a durable fact established while answering any question, or an event that changes the state of a dossier (sent, received, signed, refused, decided, filed). It goes to the file that owns that kind of fact, then its line in index.md is updated, surgically, section by section: complete the existing section, keep the rest of the file as it stands. A fact re-derived from the corpus for the third time is a wiki line that was never written. That writing is not a pass: it appends nothing to memory.jsonl, which records gestures on files alone.

An event is written twice, once in each tense. log.md gets the past: one line appended at the end, opening on the ISO date and the dossier's name so the log greps by either, and nothing in it is ever rewritten. state.md gets the present: the dossier's line is rewritten to where things stand now, who awaits whom, and the next deadline. A state line that grows a second sentence of history is a log entry in the wrong file.

index.md is a table of contents, not a digest. One line per file — - [name](file.md) — what it answers — and the hook names the question the file settles, never the answer to it. Never restate in the index a fact the target file carries: two places then own one fact, the index costs the price of both, and neither defers to the other. Past ~150 characters a line has stopped pointing and started summarising; that is the bound, and it is the whole discipline.

A replaced fact is deleted where it lived. What replaces it takes its place, and the arbitration that moved it goes to decisions.md — that is the trace it leaves. A superseded fact left standing is a second answer to a settled question with nothing marking it stale, and the reader cannot tell which of the two is current. Only a contradiction you cannot settle stays: recorded WITH its contradiction and with what would settle it, in the file, never in the index.

Done when the fact lives in one file, once, and the index still points at every file.

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.