Query a local LLM Wiki HTTP API for semantic search and graph traversal.
npx skills add https://github.com/nashsu/llm_wiki_skill --skill llm-wiki-nashsu
Please help me install this Agent Skill. Skill: llm-wiki Source: https://github.com/nashsu/llm_wiki_skill/tree/main Command: npx skills add https://github.com/nashsu/llm_wiki_skill --skill llm-wiki-nashsu
This skill bridges the gap between your AI agent and your private, locally-hosted LLM Wiki, allowing you to ground AI responses in your own curated knowledge base without exposing data to the cloud.
Ask the agent to search your LLM Wiki for information regarding a specific topic like rotary position embeddings.
High-intent search queries and answers about installing and using this skill.
You can query a local knowledge base by connecting to a locally-running LLM Wiki instance via its HTTP API. This enables semantic search, graph traversal, and source re-indexing to retrieve and synthesize private knowledge.
Yes, you can ground AI agent responses in your private wiki without cloud exposure. The skill queries a locally-hosted LLM Wiki via its built-in HTTP API, keeping all data and communication on your local machine.
To start querying your LLM Wiki knowledge base, you need a running LLM Wiki desktop application on localhost port 19828. You also need an active API token configured for secure local HTTP communication.
To perform a hybrid search across your local wiki pages, you ask the agent to search your LLM Wiki for a specific topic. The skill executes combined keyword and vector-based semantic searches via the local HTTP API.
You can update your knowledge base index by triggering backend rescans of your source documents. The skill facilitates source management by instructing the LLM Wiki application to re-index your updated files.
The best way to navigate connections between entities in a knowledge graph is by using graph traversal features. The skill queries your local LLM Wiki to explore relationships between concepts, entities, and wiki pages.
MuchunpengQuery a local SearXNG instance for privacy-focused web, image, and news searches.
itallstartedwithaideaIngest private documents into a vector index and generate cited answers.
dotlab-hqBuild RAG pipelines that ingest, chunk, embed, and retrieve documents for LLMs.
ceedaragentsSearch the web and read pages to gather current information and sources.
