agentleFS
PratiyushPratiyush/llm-wiki/.claude/skills/llmwiki-ingest/SKILL.md

llmwiki-ingest

Skill383 starsChanged 5 months agoMITView on GitHub

Ingest one source document (or a folder of them) into the llmwiki. Use when the user drops a new markdown file, PDF, or URL into the wiki and asks you to process it. The user will typically say "ingest this", "add this to the wiki", "process this file into the wiki", or point at a file under `raw/`.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 2 are distinctive, and 9 are 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.

  • 9

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Follow the Ingest Workflow from the repo's CLAUDE.md exactly:

    line 27
  2. Distinctive:

    Read the source file(s) with the Read tool

    line 29
  3. Very specific:

    Read wiki/index.md and wiki/overview.md for context

    line 30
  4. Very specific:

    Write wiki/sources/<slug>.md using the Source Page Format

    line 31
  5. Very specific:

    Update wiki/index.md — new entry under ## Sources

    line 32
  6. Very specific:

    Update wiki/overview.md if substantial new info

    line 33
  7. Very specific:

    Create/update entity pages (wiki/entities/<TitleCase>.md)

    line 34
  8. Very specific:

    Create/update concept pages (wiki/concepts/<TitleCase>.md)

    line 35
  9. Very specific:

    When the source is under raw/sessions/ (a session transcript converted by the converter):

    line 42
  10. Very specific:

    No silent overwrites. Conflicting claims go under ## Contradictions.

    line 54
  11. Very specific:

    Every page has a ## Connections section with at least one [[wikilink]].

    line 55

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

llmwiki-ingest

What this skill does

Takes a source file (or a folder) and turns it into wiki pages following the Karpathy LLM Wiki pattern:

  1. Writes a wiki/sources/<slug>.md summary
  2. Creates/updates entity pages for people, companies, projects, tools, libraries mentioned
  3. Creates/updates concept pages for ideas, patterns, decisions discussed
  4. Cross-links everything with [[wikilinks]]
  5. Flags contradictions
  6. Updates wiki/index.md, wiki/overview.md, wiki/log.md

When to use

  • User says "ingest this file", "add this to the wiki", "process this into the wiki"
  • User runs the /wiki-ingest slash command
  • User says "sync the wiki" — in that case, the llmwiki-sync skill runs the converter first, then invokes this skill for each new file

Workflow

Follow the Ingest Workflow from the repo's CLAUDE.md exactly:

  1. Read the source file(s) with the Read tool
  2. Read wiki/index.md and wiki/overview.md for context
  3. Write wiki/sources/<slug>.md using the Source Page Format
  4. Update wiki/index.md — new entry under ## Sources
  5. Update wiki/overview.md if substantial new info
  6. Create/update entity pages (wiki/entities/<TitleCase>.md)
  7. Create/update concept pages (wiki/concepts/<TitleCase>.md)
  8. Cross-link with [[wikilinks]] under ## Connections
  9. Flag contradictions under ## Contradictions
  10. Append to wiki/log.md: ## [YYYY-MM-DD] ingest | <title>

Session-specific rules

When the source is under raw/sessions/ (a session transcript converted by the converter):

  • Trust the frontmatter as authoritative (project, started, model, tools_used, etc.)
  • Do not copy the ## Conversation section verbatim — use it as raw material to summarise
  • Create a project entity page at wiki/entities/<ProjectSlug>.md with a ## Sessions list
  • Extract decisions into wiki/concepts/ — anything the user explicitly locked
  • Extract tools used — every entry in tools_used is a candidate entity
  • If is_subagent: true — link to the parent session rather than creating a new project entity

Hard rules

  1. raw/ is immutable. Never modify files there.
  2. No silent overwrites. Conflicting claims go under ## Contradictions.
  3. Every page has a ## Connections section with at least one [[wikilink]].
  4. Frontmatter is authoritative. Always populate title, type, tags, sources, last_updated.

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.