Skeletongraph is a token-minimal, constraint-preserving context assembly library for AI coding agents. It uses AST, tree-sitter, and graph techniques to optimize LLM context while maintaining code structure. The MIT-licensed Python package is designed for developers creating efficient coding assistants and agents.
skeletongraph sits in PulseGate's Developer Tools category. AI coding agents exceeding context windows or losing structural constraints when assembling large codebases. It is built as an open-source project for developers building AI coding agents. The project is open source (MIT). It ships for the command line.
It is developed by Yash Doke, and it first shipped in 2026. Development happens publicly on GitHub with 234 commits in the last 90 days. Key capabilities include context optimization, AST parsing, and token reduction. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do