Evaluate GitHub Copilot Chat in real VS Code workspaces. A Vally executor for testing AI agent skills, custom instructions, and MCP tools, with optional HAR/ATIF execution traces.
testing benchmarking automation mcp vscode visual-studio-code developer-tools ai-agents dev-proxy github-copilot ai-evaluation evals llm-evaluation github-copilot-chat model-context-protocol agent-skills agent-evaluation vally atif
-
Updated
Oct 29, 2026 - TypeScript