CowAgent knowledge-wiki
Manage the personal knowledge wiki. Use when the user shares articles, documents, or asks to organize knowledge; when a conversation produces insights worth preserving as structured knowledge; or when the user asks about the knowledge base.
git clone https://github.com/zhayujie/CowAgent
T=$(mktemp -d) && git clone --depth=1 https://github.com/zhayujie/CowAgent "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/knowledge-wiki" ~/.claude/skills/zhayujie-cowagent-knowledge-wiki && rm -rf "$T"
skills/knowledge-wiki/SKILL.mdKnowledge Wiki
Maintain a persistent, structured knowledge base in the
knowledge/ directory.
Core Operations
1. Ingest — User shares an article, document, or resource
- Read and understand the source material
- Extract key facts, insights, and structured knowledge
- Determine the appropriate subdirectory:
- Read
to see existing categoriesknowledge/index.md - If a matching category exists, follow that structure
- If not, create a new subdirectory with a clear name
- Read
- Create the knowledge page:
knowledge/<category>/<slug>.md - Update
and append toknowledge/index.mdknowledge/log.md
2. Synthesize — Conversation produces valuable structured knowledge
- Create a knowledge page under the appropriate category
- Update related pages with cross-references
- Update
andknowledge/index.mdknowledge/log.md
3. Query — User asks about accumulated knowledge
- Check
(already in your context) for relevant pagesknowledge/index.md - Read specific pages with the
toolread - Supplement with
if neededmemory_search
Page Format
# Page Title > Source: <URL or description of the original material> Content here. Cross-reference related pages with markdown links: [Related Page](../category/related-page.md) ## Key Points - ... ## Related - [Page A](../category/page-a.md) — how it relates - [Page B](../category/page-b.md) — how it relates
The
> Source: line records where the knowledge came from (URL, document name, conversation, etc.). Always include it when the material originates from a specific source.
Cross-references build a knowledge graph. When creating or updating a page, link to related pages and update those pages to link back. Only link to pages that already exist — if a concept deserves its own page, create it first, then add the link.
Index Format (knowledge/index.md
)
knowledge/index.mdFlat list, one line per page:
[Title](path) — one-line summary. Group by category (matching subdirectories). No tables, no emoji.
# Knowledge Index ## Category A - [Page Title](category-a/page-slug.md) — one-line summary ## Category B - [Page Title](category-b/page-slug.md) — one-line summary
Category names and structure are flexible — follow whatever organization already exists in the index, or create new categories based on the content.
Log Format (knowledge/log.md
)
knowledge/log.mdAppend-only, newest at bottom:
## [YYYY-MM-DD] ingest | Page Title ## [YYYY-MM-DD] synthesize | Page Title
Guidelines
- File naming: lowercase kebab-case (e.g.
)machine-learning.md - One topic per page: link between pages rather than duplicating
- Update, don't duplicate: if a page exists, update it
- Cross-reference: every page should link to related pages; keep the knowledge graph connected
- Index is mandatory: always update
after any changeknowledge/index.md - Be concise: capture essence, not copy entire sources
- Full paths in replies: when referencing knowledge files in conversation replies, use the full path from workspace root (e.g.
), not relative paths. Relative paths are only for cross-references inside knowledge pages themselves.[Title](knowledge/<category>/<slug>.md) - Cite sources: when answering based on knowledge pages, include links to the relevant pages so the user can explore further.