semhood
If you're tired of your AI agent burning context on grep and read_file calls that miss the intent of "how do we retry failed payments," this server gives Claude semantic search over your actual AST. It indexes functions and classes with Tree-sitter, embeds them alongside optional LLM-generated logic summaries and predicted developer questions, then exposes search, find_symbol, and get_chunk_context through MCP. Every result includes its call graph,what it calls and what calls it,so your agent gets the neighborhood in one shot instead of chasing references by hand. Works offline with local embeddings or plug in Voyage or OpenAI for sharper retrieval. Supports ten languages including Python, TypeScript, Go, and Rust.
- GitHub stars
- 1
- tools
- 8
- transport
- STDIO
- last updated
- Jun 2, 2026
Tools
8 toolsPublic tool metadata for what this MCP can expose to an agent.
searchSemantic search over the index.
Semantic search over the index.
No parameter schema in public metadata yet.
search_manySeveral searches in one call — results grouped per query.
Several searches in one call — results grouped per query.
No parameter schema in public metadata yet.
find_symbolExact-name lookup for a function/method/class.
Exact-name lookup for a function/method/class.
No parameter schema in public metadata yet.
get_chunk_contextFull source + calls + called_by for a symbol.
Full source + calls + called_by for a symbol.
No parameter schema in public metadata yet.
index_statusSanity-check the index.
Sanity-check the index.
No parameter schema in public metadata yet.
list_pending_enrichmentsGet a batch of chunks needing LLM enrichment.
Get a batch of chunks needing LLM enrichment.
No parameter schema in public metadata yet.
save_enrichmentPersist a summary the agent wrote.
Persist a summary the agent wrote.
No parameter schema in public metadata yet.
enrichment_progressLoop sentinel — pending vs done.
Loop sentinel — pending vs done.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
semhood
Stop grepping. Find the exact code your AI agent needs by intent, not keywords. semhood is an AST-aware semantic code search engine that retrieves code by what it does, complete with call-graph context and optional LLM enrichment.
Runs fully offline with zero API keys — or plug in cloud embeddings — Voyage's code-specialized models, or OpenAI's strong general-purpose (natural-language) embeddings — for higher-quality retrieval. Optional LLM enrichment adds a logic summary and developer queries to each chunk that you commit once and share — and every result ships with its call graph (what it calls + what calls it).
Details
- Repository
- github.com/ahmeedgamil/semhood
- Category
- Search & Web Crawling
- Package
semhood- Transport
- STDIO
- Registry
- MCP registry, active
Related Search & Web Crawling MCP servers
View all
Wikivibe MCPalexmi64/wikivibe-mcpSearch and read public Wikivibe articles about AI coding, agents, MCP, GEO, bots and deployment.1stars
Scholardatacloudcorprecords/scholardataGoogle Scholar as JSON: scholarly search results, case-law opinions and citation formats.1stars
LinkDRillyism/linkdr-mcpBacklink research, domain authority, SERP analysis, site approvals, and link building for agents.1stars
Sciganticscigantic/scigantic-mcpSearch public scientific datasets and read schema cards that describe their files, columns and rows1stars
1staystayker-com/1stay-mcpHotel booking MCP server. Search, book, and manage reservations across 250K+ properties worldwide.1stars