EtienneChollet/ontomics
by Various
Extract domain knowledge from codebases to reduce LLM token consumption by 20x and time in agentic search by 10x — gathers and makes concepts, naming conventions, and vocabulary qu
MCP
EtienneChollet/ontomics
Added 1 June 2026
Overview
Ontomics extracts domain knowledge from codebases by analyzing concepts, naming conventions, and vocabulary, then makes them queryable via MCP. It reduces LLM token consumption by 20x and agentic search time by 10x by providing a structured ontology instead of raw code.
Best for
Best for
Developers working with large codebases who want to reduce LLM costs and speed up agentic search
Use cases
- Pre-process a codebase to build a queryable ontology for LLM-assisted development
- Reduce token costs when using LLMs to understand or refactor large codebases
- Speed up agentic code search by replacing raw code scans with ontology lookups
How to use
Install
npm install -g @ontomics/ontomics Tools exposed
query_conceptlocate_conceptdescribe_symboltrace_conceptlist_conceptslist_conventionslist_entitiescheck_namingsuggest_namevocabulary_healthontology_diffexport_domain_packfind_similar_logicdescribe_logiccompact_contextdescribe_fileconcept_maptype_flowstrace_type
Tested with
Claude Code, ChatGPT
Notes
Ontomics extracts domain knowledge from codebases by analyzing concepts, naming conventions, and vocabulary, then makes them queryable via MCP. It reduces LLM token consumption by 20x and agentic search time by 10x by providing a structured ontology instead of raw code.
35 stars on GitHub. Last updated 2026-04-17. Licensed MIT.
Use cases
- Pre-process a codebase to build a queryable ontology for LLM-assisted development
- Reduce token costs when using LLMs to understand or refactor large codebases
- Speed up agentic code search by replacing raw code scans with ontology lookups
Pros
- Written in Rust for performance and safety
- Delivers measurable reductions in token usage and search time
- Exposes domain knowledge via MCP for easy integration with LLM tools
Cons
- Limited community adoption with only 35 GitHub stars
- Requires initial setup and understanding of MCP protocol
- Effectiveness depends on codebase structure and naming consistency
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Written in Rust for performance and safety
- Delivers measurable reductions in token usage and search time
- Exposes domain knowledge via MCP for easy integration with LLM tools
Cons
- Limited community adoption with only 35 GitHub stars
- Requires initial setup and understanding of MCP protocol
- Effectiveness depends on codebase structure and naming consistency
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
oraios/serena
Various
A powerful MCP toolkit for coding, providing semantic retrieval and editing capabilities - the IDE for your agent
repowise-dev/repowise
Various
Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.
