An MCP server that provides intelligent semantic code search for AI assistants using local AI models.
smart-coding-mcp is worth checking the docs before setup with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
gh repo view omar-haris/smart-coding-mcp --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
This tool helps AI coding assistants find relevant code by meaning, not just keywords. It indexes your codebase with AI embeddings so you can search conceptually, like 'where is authentication handled?' and find related code even if it uses different terms. It runs locally, keeping your code private.
Smart Coding MCP is an extensible Model Context Protocol (MCP) server that enhances AI coding assistants with intelligent semantic code search. It uses local AI models with Matryoshka Representation Learning (MRL) to generate flexible embedding dimensions (64-768d), enabling searches based on meaning rather than exact keywords. The server indexes your codebase into a SQLite cache, allowing fast, hybrid search (cosine similarity + exact match boosting). It includes tools for semantic search, package version lookup, manual reindexing, cache clearing, and workspace switching. Designed for privacy, all processing happens locally. It supports multiple AI assistants like Claude, Codex, Gemini, and Cursor, and integrates with any MCP-compatible client.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 152 days ago.
198 GitHub stars indicate community interest.
5 open issues signal maintenance load.
MIT license detected.
Find code related to authentication even when terms like 'login' or 'session' are used.
Explore unfamiliar codebases by asking conceptual questions like 'How does error handling work?'
Check latest package versions across 20+ ecosystems before adding dependencies.
Reindex codebase after major refactoring to keep search results accurate.
Switch between multiple projects in a monorepo without restarting the server.
Embedding models may consume significant CPU/GPU resources during indexing.
Large codebases may require substantial disk space for the cache (SQLite + embeddings).
The tool runs locally, but if used with cloud-based AI assistants, code snippets may be sent to those services.
198
Stars
31
Forks
5
Issues
MIT
License
Local codebase intelligence CLI and MCP server for AI coding agents with change-safety gates and audit evidence.
An offline MCP server that indexes your codebase for semantic search, code search, and git history retrieval.
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
3 security/trust notes recorded.
Setup difficulty is 3/5.