Saucebox Registry
← Browse

@ybcarry/llm-wiki

B

---

skillclaude

Install

agr install @ybcarry/llm-wiki --target claude

Writes 4 files into .claude/skills/, pinned to git-4194c802.

  • .claude/skills/llm-wiki/README.en.md
  • .claude/skills/llm-wiki/README.md
  • .claude/skills/llm-wiki/README.zh.md
  • .claude/skills/llm-wiki/SKILL.md

Document

name: llm-wiki description: Maintains a persistent markdown wiki from raw sources. Ingests raw notes, compiles structured wiki pages, updates index and log files, and produces question-answer outputs. Use for LLM research notes, knowledge bases, source-to-wiki compilation, topic synthesis, and wiki maintenance.

llm-wiki

Purpose

Maintain a persistent, compounding markdown wiki built from raw sources.

When to use

  • User wants to organize research sources into a wiki
  • User asks to compile raw notes into structured knowledge pages
  • User wants Q&A, comparisons, or summaries based on the wiki
  • User wants the wiki index or log updated after ingesting sources

Core workflow

  1. Put source material in raw/.
  2. Read raw sources and integrate them into wiki/.
  3. Update index.md and log.md.
  4. Write question-based analysis into outputs/.

File layout

  • raw/ — immutable source material
  • wiki/ — structured knowledge pages
  • outputs/ — Q&A and analysis based on the wiki
  • index.md — content-oriented catalog
  • log.md — chronological activity log
  • schema.md — operating conventions and page templates

Maintenance rules

  • Keep raw sources unchanged unless explicitly asked to archive or copy.
  • Prefer modular wiki pages with cross-references.
  • When sources contradict prior claims, update the wiki and note the conflict.
  • Mark uncertain claims with [待验证].
  • Every ingest should update the index and log.

Output style

  • Use concise markdown.
  • Favor source-linked synthesis over generic summaries.
  • Preserve traceability from wiki pages back to raw sources.

References

  • references/creation-guide.md — skill creation rationale
  • references/skill-structure.md — optional structure notes
  • references/best-practices.md — writing and maintenance guidance

Repository README

Describes YBCarry/llm-wiki as a whole, which may contain artifacts other than this one. Where this artifact had no useful description of its own, its summary was taken from here.

llm-wiki

This repository has bilingual README files:

The project idea comes from Karpathy's llm-wiki.md: https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f

The repository implements a three-layer workflow:

  • raw/ — source material
  • wiki/ — structured knowledge pages
  • outputs/ — Q&A and analysis outputs

See schema.md for the maintenance rules and page templates.

Trustgrade B

  • passBody integrity

    Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.

  • passType matchnot applicable to this artifact type

    Whether the artifact is really the kind of thing its metadata claims it is.

  • passFreshness

    How long since the source repository was last pushed to.

  • passPrompt injection

    Scans the artifact's own text for instructions aimed at your agent rather than at you.

  • warnLicenseno SPDX license detected

    Whether the source repository declares an SPDX license permissive enough to redistribute.

How the grade is calculated

Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:

  • Aevery check passed
  • Bone warning
  • Ctwo warnings
  • Dprompt injection or body integrity failed, or three warnings
  • Fone of those failed, and something else is wrong

These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.

Versions

  • git-4194c80249362026-07-31