context-sherpa: Context Sherpa is a platform for Context Engineering that enhances AI coding agents by providing precise symbolic signals from codebases. It uses SCIP-based indexing and structural analysis to reduce token consumption by up to 90% while improving accuracy. The tool offers both a GUI Code Atlas Explorer and a headless MCP server for integration with tools like Cursor and Cline.; xLAM: xLAM is a research repository for Large Action Models (LAMs), which aggregates and unifies agent trajectories from diverse environments into a consistent format. It streamlines the creation of a generic data loader optimized for agent training, enabling robust model development across various scenarios.
Enhancing AI coding agents with precise codebase signals
Function calling in LLMs