Search and retrieve relevant context from local files using semantic search powered by LanceDB. For developers building AI applications that need intelligent document retrieval without cloud dependencies.
io.github.cornelcroi/context-lens
Local install
STDIO
No auth required
How models use it and what it is built for.
Search and retrieve relevant context from local files using semantic search powered by LanceDB. For developers building AI applications that need intelligent document retrieval without cloud dependencies.
Local install — runs as a subprocess.
Where to find authoritative docs and source for Context Lens.
Paste any of these into Agent Studio after connecting Context Lens.
Common questions about connecting and running Context Lens.
What file types does context-lens support?
The server supports 25+ file types for semantic indexing and search. Exact supported formats are documented in the project repository; common types include markdown, JSON, PDFs, and code files.
Do I need to set up a database or cloud service?
No. Context-lens uses LanceDB and runs 100% locally with zero setup required. All indexing and search happen on your machine.
How do I install and run context-lens?
Install via `uvx context-lens@0.1.8`. It communicates over stdio, so it integrates directly with MCP-compatible clients without additional configuration.
Is there a cost to use context-lens?
No. Context-lens is open-source and runs entirely locally, so there are no API fees or subscription costs.
What's the difference between context-lens and other semantic search tools?
Context-lens is designed for local-first development with no cloud dependency, supports 25+ file types out of the box, and integrates as an MCP server for seamless AI assistant integration.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 30+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent Studio