Answers questions from a local Obsidian-style wiki by reading its index first, then the relevant pages, and replying with wiki-link citations instead of model memory.

No licenceAuto-check passedKnowledge Management

SKILL.md written in Chinese; this summary is our English description.

Install Wiki Vault Query

skills CLI
$ npx skills add jason-effi-lab/karpathy-llm-wiki-vault --skill query -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jason-effi-lab/karpathy-llm-wiki-vault query --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jason-effi-lab/karpathy-llm-wiki-vault.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/query .claude/skills/query && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
query
GitHub stars
720
Token cost
~383 tokens
SKILL.md length
97 words
Files
1
Skills in repo
3
Repo updated
First seen
Licence
None found

At a glance

Answers questions from a local Obsidian-style wiki by reading its index first, then the relevant pages, and replying with wiki-link citations instead of model memory.

  • Asking what your notes say about a topic
  • SKILL.md covers 核心目标, 触发场景, 降级策略 and 检索与综合流水线, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Looking up a past decision recorded in the wiki

What it does

The skill triggers on /query or on natural questions about your notes, past decisions or knowledge base. The agent must start by reading wiki/index.md to find relevant entities, concepts, sources and syntheses, then read the best pages in full and answer with double-bracket page citations, quoting original text in block quotes and citing a page once at the start and end of a paragraph.

Answers from model memory are forbidden. If the wiki has nothing relevant, the agent has to say that the local knowledge base had no match before giving a general answer. A substantial answer, one longer than two paragraphs or comparative in nature, prompts a question about saving it as a synthesis page under wiki/syntheses, registered in the index, and every query ends with an entry appended to wiki/log.md in the format set by CLAUDE.md. The SKILL.md is written in Chinese.

When your agent uses it

  • Asking what your notes say about a topic
  • Looking up a past decision recorded in the wiki
  • Getting an answer that cites wiki pages with double-bracket links
  • Saving a valuable multi-paragraph answer as a synthesis page

Example prompts

  • “/query my notes on the caching decision”
  • “Look up what I decided last quarter about vendor selection in my knowledge base.”
  • “Summarize what the wiki says about onboarding and offer to save it as a synthesis.”

Requirements

  • A local wiki folder with wiki/index.md, wiki/log.md and a CLAUDE.md schema

What it can do on your machine

Read from SKILL.md and the folder at commit 18f4e71. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Wiki Vault Query loads about 383 tokens when it runs. Until then it costs about 44 tokens; SKILL.md has 97 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~44
When it runs · the whole SKILL.md, loaded when a task matches
~383

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Without a licence we can't republish the file, so here is its outline and opening line. It has 97 words (~383 tokens).

name
query
user-invocable
true

Read the full SKILL.md on GitHub

Files

Just SKILL.md in .claude/skills/query of jason-effi-lab/karpathy-llm-wiki-vault.

Open the folder on GitHubat commit 18f4e71

Compare with similar skills

Wiki Vault Query next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Wiki Vault Query compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Wiki Vault Query this skilljason-effi-lab/karpathy-llm-wiki-vault720—~383Automated safety check: PassNone
LLM Wikilewislulu/llm-wiki-skill655—~3.7kAutomated safety check: PassNone
LLM WikiCharlesHoskinson/sevenlayer116—~1.5kAutomated safety check: PassCustom licence
LLM Wikipraneybehl/llm-wiki-plugin118—~5.7kAutomated safety check: PassMIT
My LLM WikiMartinLwx/dotfiles140—~2.7kAutomated safety check: PassNone
Obsidian Knowledge BuilderMichael-OvO/obsidian-knowledge-agent205—~917Automated safety check: PassMIT

Similar skills

  • LLM Wiki

    lewislulu/llm-wiki-skill

    Build and maintain a Karpathy-style LLM knowledge base — a self-compiling Obsidian markdown wiki where an Agent ingests raw sources, compiles cross-linked concept/entity/summary pages, answers…

    655 GitHub stars~3.7k tokensUpdated 5 mo ago
    Knowledge ManagementAuto-check passed
  • LLM Wiki

    CharlesHoskinson/sevenlayer

    Builds and maintains a persistent, interlinked Obsidian-compatible markdown knowledge wiki in a git repo via three operations — ingest a source into linked pages, answer a question from the…

    116 GitHub stars~1.5k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check passed
  • LLM Wiki

    praneybehl/llm-wiki-plugin

    Build and maintain an LLM-curated knowledge base from papers, articles, transcripts, notes and project findings.

    118 GitHub stars~5.7k tokensUpdated 26 days ago
    Knowledge ManagementAuto-check passed
  • My LLM Wiki

    MartinLwx/dotfiles

    Provides access to the user's personal wiki, including notes, research, project documentation, decisions, and archived knowledge.

    140 GitHub stars~2.7k tokensUpdated today
    Knowledge ManagementAuto-check passed
  • Obsidian Knowledge Builder

    Michael-OvO/obsidian-knowledge-agent

    Turns PDFs, slides, syllabi, papers and transcripts into structured, teaching-quality Obsidian notes, matching how much structure is built to how much the material needs.

    205 GitHub stars~917 tokensUpdated 3 mo ago
    Knowledge ManagementAuto-check passed
  • Turn a folder of Markdown notes (Obsidian vault, Notion export, plain repo docs) into a navigable static HTML knowledge base bundled as a single .zip file.

    210 GitHub stars~1.4k tokensUpdated 16 days ago
    Knowledge ManagementAuto-check passed

More from jason-effi-lab/karpathy-llm-wiki-vault

  • Obsidian Vault Ingest Pipeline

    jason-effi-lab/karpathy-llm-wiki-vault

    Compiles raw articles, papers and transcripts sitting in an inbox folder into linked Obsidian source, entity and concept pages, then archives the originals.

    720 GitHub stars~669 tokensUpdated 5 mo ago
    Auto-check passed
  • Knowledge Base Health Lint

    jason-effi-lab/karpathy-llm-wiki-vault

    Scans an Obsidian-style wiki for broken double-link references, orphan pages, files missing from the index and unresolved knowledge conflicts, then reports the findings.

    720 GitHub stars~336 tokensUpdated 5 mo ago
    Auto-check passed

Works with

Questions about Wiki Vault Query

What does Wiki Vault Query do?

Answers questions from a local Obsidian-style wiki by reading its index first, then the relevant pages, and replying with wiki-link citations instead of model memory. The skill triggers on /query or on natural questions about your notes, past decisions or knowledge base.md to find relevant entities, concepts, sources and syntheses, then read the best pages in full and answer with double-bracket page citations, quoting original text in block quotes and citing a page once at the start and end of a paragraph.

When should I use Wiki Vault Query?

Wiki Vault Query fits situations like: asking what your notes say about a topic; looking up a past decision recorded in the wiki; getting an answer that cites wiki pages with double-bracket links; saving a valuable multi-paragraph answer as a synthesis page.

How do I install Wiki Vault Query in Claude Code?

Run `npx skills add jason-effi-lab/karpathy-llm-wiki-vault --skill query -a claude-code`. Or copy the skill folder (.claude/skills/query in jason-effi-lab/karpathy-llm-wiki-vault) into .claude/skills/query in your project. Claude Code loads it when a task matches its description.

How do I install Wiki Vault Query in Codex?

Run `npx skills add jason-effi-lab/karpathy-llm-wiki-vault --skill query -a codex`. Or copy the skill folder (.claude/skills/query in jason-effi-lab/karpathy-llm-wiki-vault) into .agents/skills/query in your project. Codex loads it when a task matches its description.

Can I use Wiki Vault Query in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jason-effi-lab/karpathy-llm-wiki-vault --skill query -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/query, .gemini/skills/query, .github/skills/query and .opencode/skills/query in your project.

What does Wiki Vault Query need to run?

SKILL.md names no scripts, command-line tools or credentials: Wiki Vault Query is instructions for the agent only. Our summary lists: A local wiki folder with wiki/index.md, wiki/log.md and a CLAUDE.md schema.

Does Wiki Vault Query access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Wiki Vault Query safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Wiki Vault Query use?

No licence was found for Wiki Vault Query or its repository. Without one, default copyright applies: ask the author before reusing or redistributing it.

How many tokens does Wiki Vault Query use?

About 383 tokens (SKILL.md is roughly 1.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Wiki Vault Query?

Skills that share tags, products or a category with Wiki Vault Query: LLM Wiki (lewislulu/llm-wiki-skill, 655 stars), LLM Wiki (CharlesHoskinson/sevenlayer, 116 stars), LLM Wiki (praneybehl/llm-wiki-plugin, 118 stars) and My LLM Wiki (MartinLwx/dotfiles, 140 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Wiki Vault Query?

jason-effi-lab (a GitHub user) maintains it in jason-effi-lab/karpathy-llm-wiki-vault, which has 720 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on April 13, 2026.

Source: jason-effi-lab/karpathy-llm-wiki-vault on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.