bborbe/semantic-search-mcp
If you are the rightful owner of semantic-search-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.
Semantic Search MCP is a server designed for semantic search over markdown files, utilizing sentence-transformers for embeddings and FAISS for vector search.
Semantic Search MCP
MCP server that adds semantic search over markdown files to Claude Code. Find related notes by meaning, not just keywords. Detect duplicates before creating new notes.
Installation
From GitHub:
claude mcp remove -s project semantic-search
claude mcp add -s project semantic-search \
--env CONTENT_PATH=/path/to/your/content \
-- \
uvx --from git+https://github.com/bborbe/semantic-search-mcp semantic-search-mcp serve
Local development:
claude mcp remove -s project semantic-search
claude mcp add -s project semantic-search \
--env CONTENT_PATH=/path/to/your/content \
-- \
uvx --reinstall --from /path/to/semantic-search-mcp semantic-search-mcp serve
Replace /path/to/your/content with your markdown directory (e.g., Obsidian vault path).
Usage
Ask Claude:
- "Do I have notes about X?" → searches related notes
- "Find similar notes to Y" → finds semantically related content
- Before creating new notes → Claude checks for duplicates automatically
Tools
search_related(query, top_k=5)- Find semantically related notescheck_duplicates(file_path)- Detect duplicate/similar notes
How It Works
First run downloads a small embedding model (~90MB) and indexes your markdown files (<1s for typical vaults). Each Claude Code session gets its own index in /tmp/ that auto-updates when files change. Multiple sessions work independently without conflicts.
License
This project is licensed under the BSD 2-Clause License - see the file for details.