← 返回 Skills 市场
546
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install llm-wiki-skill
功能描述
LLM-maintained knowledge wiki for Obsidian vaults. The LLM does the bookkeeping, you do the curation. Three sub-skills: ingest sources into the wiki, query a...
使用说明 (SKILL.md)
/llmwiki — LLM-Maintained Knowledge Wiki
You are a knowledge management assistant. Your job is to route the user's request to the correct sub-skill based on their intent, then execute it.
Trigger
User invokes /llmwiki followed by a sub-command and optional arguments:
/llmwiki ingest my-note.md
/llmwiki ingest https://example.com/article
/llmwiki query what do I know about machine learning
/llmwiki health
Routing
Parse the first word of $ARGUMENTS and delegate:
| Sub-command | Delegates to | Description |
|---|---|---|
ingest \x3Csource> |
skills/llmwiki-ingest/SKILL.md |
Extract insights from a file, URL, or topic into _wiki/ |
query \x3Cquestion> |
skills/llmwiki-query/SKILL.md |
Search the wiki and synthesize an answer |
health |
skills/llmwiki-health/SKILL.md |
Audit wiki for orphans, stubs, broken links, stale pages |
If no sub-command is given, print this help:
Usage:
/llmwiki ingest \x3Cfile|url|topic> — add knowledge to the wiki
/llmwiki query \x3Cquestion> — ask the wiki a question
/llmwiki health — audit wiki quality
Use Cases
- Build a compounding personal knowledge base from Obsidian notes
- Capture insights from articles and documentation without losing them
- Retrieve past research without re-reading raw notes
- Keep the wiki well-connected and free of stale content
References
安全使用建议
This skill appears to do exactly what it says: read files/URLs and create/update an Obsidian _wiki. Before installing, do these simple checks: 1) Back up your vault (so auto-writes are recoverable). 2) Be aware the skill will append to and edit files (_wiki/ and log.md) and may auto-add wikilinks or fill frontmatter — if you prefer manual edits, disable autonomous invocation or run it in a test copy first. 3) Confirm you are comfortable with the agent having filesystem and web-fetch (WebFetch) access, since ingesting URLs will fetch external pages. 4) If you want tighter control, run /wiki-ingest and /wiki-health only interactively rather than allowing background/autonomous runs. If you want more assurance, ask the author for an explicit description of what tooling (Read, Grep, WebFetch) the platform will grant to the skill at runtime and test the skill on a small sandbox vault first.
功能分析
Type: OpenClaw Skill
Name: llm-wiki-skill
Version: 1.0.0
The llm-wiki-skill bundle is a legitimate set of tools for managing a knowledge wiki within an Obsidian vault. It provides sub-skills for ingesting content from local files or URLs (llmwiki-ingest), querying the synthesized knowledge (llmwiki-query), and auditing the wiki for quality issues like broken links or orphans (llmwiki-health). The instructions provided to the AI agent in the SKILL.md files are well-structured and strictly aligned with the stated purpose of knowledge synthesis and maintenance. While the skills utilize powerful capabilities such as WebFetch and filesystem access, these are used transparently and appropriately for the described functionality.
能力评估
Purpose & Capability
Name and description (LLM-maintained Obsidian wiki) match the actual behavior: the skill reads vault files, fetches web pages, searches (grep/glob), and creates/updates markdown pages in _wiki/. No unrelated env vars, binaries, or external credentials are requested.
Instruction Scope
Runtime instructions explicitly direct the agent to read arbitrary files in the vault, Grep the full vault, WebFetch URLs, Glob _wiki/**/*.md, and create or modify _wiki pages and _wiki/log.md. This is coherent with the purpose but important: the skill will autonomously write changes (including auto-fixes) to your files unless you restrict model actions.
Install Mechanism
Instruction-only skill with no install steps or downloaded code. Lowest-risk install footprint: nothing is written to disk by an installer and there are no external packages to fetch during install.
Credentials
No environment variables, credentials, or config paths are requested. The skill does require filesystem and web-fetching capabilities (Read, Grep, WebFetch) which are appropriate for ingesting files and URLs into a wiki.
Persistence & Privilege
always:false (good). The skill will modify files in your vault and auto-update the wiki on queries and auto-fix some issues on health runs. Combined with normal platform autonomous invocation, this means the skill could make changes without a separate explicit manual edit—recommend reviewing and limiting autonomous execution if undesired.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install llm-wiki-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/llm-wiki-skill触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of llmwiki skill.
- Adds an LLM-powered assistant for managing a knowledge wiki in Obsidian.
- Supports three main sub-commands: ingest (add content), query (ask questions), and health (audit quality).
- Enables automatic routing to delegate tasks based on user input.
- Provides help and usage instructions for all available commands.
- Designed to streamline personal knowledge base curation and maintenance.
元数据
常见问题
LLM-Wiki-Skill 是什么？
LLM-maintained knowledge wiki for Obsidian vaults. The LLM does the bookkeeping, you do the curation. Three sub-skills: ingest sources into the wiki, query a... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 546 次。
如何安装 LLM-Wiki-Skill？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install llm-wiki-skill」即可一键安装，无需额外配置。
LLM-Wiki-Skill 是免费的吗？
是的，LLM-Wiki-Skill 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
LLM-Wiki-Skill 支持哪些平台？
LLM-Wiki-Skill 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 LLM-Wiki-Skill？
由 Denny（@webkong）开发并维护，当前版本 v1.0.0。
推荐 Skills