mcp-rust-docs-embed
If you are the rightful owner of mcp-rust-docs-embed and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
An MCP server that provides semantic search capabilities for Rust crate documentation through vector embeddings.
The MCP (Model Context Protocol) server is designed to enhance AI assistants' ability to search and comprehend Rust crate documentation by leveraging vector embeddings. It automates the generation of JSON-formatted documentation using the Rust nightly toolchain and creates vector embeddings of the content with OpenAI's technology. These embeddings are stored in Qdrant, a vector database, to facilitate efficient semantic searches. The server supports version management, ensuring that each crate version is stored separately, allowing for precise version-specific searches. It also intelligently selects compatible features to avoid conflicts and handles long-running embedding tasks asynchronously, providing status tracking for these operations. This setup is particularly useful for developers and AI systems that need to interact with Rust documentation in a more intuitive and natural language-driven manner.
Features
- Automatic Documentation Generation: Builds Rust documentation in JSON format using cargo's nightly toolchain.
- Semantic Search: Allows querying of documentation using natural language through vector embeddings.
- Version Management: Stores each crate version separately for precise version-specific searches.
- Feature Support: Intelligently selects compatible features to avoid mutually exclusive conflicts.
- Async Operations: Handles long-running embedding tasks asynchronously with status tracking.
Tools
embed_docs
Generate and embed documentation for a Rust crate.
query_embeddings
Search embedded documentation using natural language.
check_embed_status
Check the status of an ongoing embedding operation.
list_embedded_crates
List all crates and versions that have been embedded.