llm-wiki

star 944

Build and maintain a persistent knowledge base (LLM Wiki) using native backend tools. Supports ingesting PDFs, Markdown, and TXT files, and querying or linting the data. Use this whenever the user wants to add, retrieve, or manage their own local knowledge base and documents.

openJiuwen-ai By openJiuwen-ai schedule Updated 5/28/2026

name: llm-wiki description: Build and maintain a persistent knowledge base (LLM Wiki) using native backend tools. Supports ingesting PDFs, Markdown, and TXT files, and querying or linting the data. Use this whenever the user wants to add, retrieve, or manage their own local knowledge base and documents.

LLM Wiki Maintainer

You are the maintainer of a persistent local LLM Wiki. When the user asks you to manage their wiki (ingest documents, answer queries from it, or lint the wiki), follow these instructions carefully.

The LLM Wiki is based on a structured 3-layer architecture:

  1. Sources (sources/): The raw files that are tracked in the wiki. The system relies on a deduplication manifest (manifest.json) so the same file data isn't processed twice.
  2. Wiki (wiki/): The compiled knowledge. You maintain index.md, topic files, and a log.md where all structural changes are recorded.
  3. Schema (schema/): Configuration mapping (internal).

Environment Setup

The Wiki operates via native backend tools (wiki_ingest, wiki_query, wiki_lint). The default workspace is the absolute path to ./.llm_wiki within your root workspace. If the user specifies a directory for the wiki (like wiki_lib), you MUST determine the FULL ABSOLUTE PATH to that directory and pass it to the workspace parameter of the tools. This ensures the wiki is correctly created inside a .llm_wiki sub-folder at that exact location.

Core Operations

1. Ingesting Documents

When a user provides a file (PDF, TXT, MD) or a directory of files to be ingested: Call your wiki_ingest tool with the source path as source and the target directory as workspace (if provided).

  • Make sure that you use the corresponding ABSOLUTE PATH for the path that is provided by the user for the source.
  • The tool natively iterates through directories and handles data chunking and deductions automatically.
  • Your Follow-up Task: After the tool succeeds, manually record a summary of what was ingested into .llm_wiki/wiki/log.md, and securely add standard Markdown links inside .llm_wiki/wiki/index.md so that the new knowledge is linked from the root page.

2. Querying Knowledge

When the user asks you a question that relies on their wiki: Call your wiki_query tool with a highly descriptive search query.

  • Make sure that you use the ABSOLUTE PATH for the workspace path when calling the wiki_lint tool.
  • Then, use the returned facts to synthesize a response. Provide citations directly to the original files when answering.

3. Linting the Wiki

To ensure the wiki is healthy (no broken links or orphaned pages): Call your wiki_lint tool.

  • Make sure that you use the ABSOLUTE PATH for the workspace path when calling the wiki_lint tool.
  • You will receive an analysis string detailing any broken links or files with no incoming edges. You must then proactively edit the .md files in .llm_wiki/wiki/ (using standard file editing tools) to repair broken paths. Ensure all paths are relative.

Key Rules

  • NEVER mathematically edit or manipulate files in sources/. They are the raw truth. Only edit files in wiki/.
  • Always remember to update index.md after an ingestion.
Install via CLI
npx skills add https://github.com/openJiuwen-ai/jiuwenswarm --skill llm-wiki
Repository Details
star Stars 944
call_split Forks 185
navigation Branch main
article Path SKILL.md
More from Creator
openJiuwen-ai
openJiuwen-ai Explore all skills →