Hugoduncan Mcp Vector Search
hugoduncan-mcp-vector-search is a Claude Code plugin for the Build phase that connects agents to semantic document search through MCP embedding and server configuration.
by hugoduncan · github.com/hugoduncan/mcp-vector-search
Give Claude and other agents semantic search over your local documents through an MCP vector-search server.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install hugoduncan-mcp-vector-search@hugoduncan/mcp-vector-searchBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:hugoduncan/mcp-vector-search") and the paste-ready config with skillselion.get_install_config("plugin:hugoduncan/mcp-vector-search").
What it does
hugoduncan-mcp-vector-search is a Claude Code plugin for semantic document search delivered through the Model Context Protocol. Solo builders who want agents to retrieve relevant passages from their own files—specs, notes, code docs—can register this bundle instead of hand-rolling embeddings, index management, and MCP tool definitions. The entry is categorized as configuration and integration, reflecting embed pipelines, server setup, and writing or config guides across two plugins. It targets agent-first workflows: your coding agent queries a vector index rather than stuffing entire folders into context. Use it during Build when you are connecting RAG-like retrieval to Claude Code, Cursor, or other MCP clients. It does not replace a hosted vector database product; it packages the hugoduncan mcp-vector-search project for local or self-managed semantic search.
Highlights
- Semantic search over your files exposed to AI agents via MCP
- Embedding and configuration-oriented plugin keywords (embed, config, servers)
- Two-plugin bundle: MCP vector search plus setup/configuration guidance
- Library-style integration rather than a single opaque command
- Fits document-grounded agents without building search from scratch
Why builders use it
Agents cannot reliably find the right paragraph in your docs without a vector index and MCP tools you have not set up yet.
After install, configured MCP vector search lets agents embed and query your documents semantically during coding sessions.
At a glance
- Type - Plugin in LLM Integration.
- Adoption - 0 installs, 2 stars, 0 votes.
FAQ
Who is hugoduncan-mcp-vector-search for?
Developers using Claude Code or MCP-compatible agents who want semantic search across their own document files.
When should I use hugoduncan-mcp-vector-search?
Use it while building agent integrations when context window limits force you to retrieve relevant docs instead of pasting everything.
How do I add hugoduncan-mcp-vector-search to my agent?
Install the plugin from hugoduncan/mcp-vector-search, configure embedding and MCP server settings per the bundle guides, then point Claude Code at the MCP vector search server.
Comments
Share how you use hugoduncan-mcp-vector-search, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.