llm-corpus
v0.0.1
Published
Local-first, LLM-native knowledge base CLI toolkit. Ingest, classify, and search documents with BM25/FTS5 ranking.
Downloads
160
Maintainers
Readme
llm-corpus
Local-first, LLM-native knowledge base CLI toolkit.
Ingest documents (PDF, web, audio, video), classify with local LLM, search with BM25/FTS5 ranking. Works as a standalone CLI or as an MCP server for AI agents (Claude Code, Gemini CLI, Codex CLI).
Status: Under active development. Full release coming soon.
Planned Features
corpus search "query"— BM25-ranked full-text searchcorpus ingest --url <URL>— Ingest web pages, PDFs, audio, videocorpus validate— Schema compliance checkingcorpus mcp-server— MCP server for AI agent integration- Pluggable LLM (Ollama, LM Studio, cloud APIs, or none)
- Progressive feature tiers (search-only → full pipeline)
