About
What is SigMap?
SigMap is a zero-dependency AI context engine that reduces LLM token usage by up to 97%. It maps only the signals that matter from your codebase — functions, types, exports — across 21 languages. Includes an MCP server for direct integration with Claude, ChatGPT, and other LLM tools.
Best used for
Reducing LLM token costs, optimizing AI context windows, codebase-aware AI assistants
LLMtoken optimizationcontext windowMCPDeveloper Toolsopen-source
Capabilities
Key features
Not yet documentedTarget Audience
Software engineersAI developerssolution architects
Integrations
Not yet documentedPricing & Plans
Open-source · Starting at USD 0.00
Free and open source. MIT license.
FAQs
What is SigMap used for?
SigMap extracts only the meaningful signals from your codebase — function signatures, types, exports — and feeds them to your LLM instead of raw file contents. This reduces token usage by up to 97% with no loss in AI understanding.
Which languages and tools does SigMap support?
SigMap supports 21 programming languages and includes a built-in MCP server compatible with Claude, ChatGPT, and any LLM that supports the Model Context Protocol.
Is SigMap free to use?
Yes — SigMap is fully open source under the MIT license with zero dependencies. Available via npm.
Shypd.aiThe largest open index of AI tools
