AIAny
AI Client2025
Icon for item

vscode-copilot-chat

Source code for the GitHub Copilot Chat extension in VS Code: inline chat, an agent mode that plans and edits files autonomously, next-edit suggestions, and MCP integration. Open-sourced so developers can study how Copilot connects to an editor.

Introduction

Microsoft rarely ships the full source of a flagship commercial extension, yet here is the entire Copilot Chat codebase, archived in May 2026 once development folded back into the main VS Code repo. The real value is not the feature list everyone already knows; it's the chance to read how a production AI coding assistant is actually wired into an editor, prompt construction and all, rather than guessing from a demo.

What Sets It Apart
  • The agent loop is in the open: planning, file edits, command execution, and self-correction are coded as concrete TypeScript orchestration, not hidden behind a service boundary, so you can trace exactly how a task becomes a sequence of tool calls.
  • It is model-agnostic in practice. Beyond Copilot's own service models, the code shows how Claude and Codex providers are slotted in, which means the context-assembly logic is decoupled from any single LLM.
  • MCP server integration and custom instructions are first-class, so the extension doubles as a worked example of grounding an editor agent in external tools and project-specific rules.
Who It's For

Great fit if you are building an editor-integrated coding agent and want a real reference for prompt assembly, context gathering, and tool dispatch at production scale. Look elsewhere if you want something runnable or forkable: the repo is archived, active work moved to microsoft/vscode, releases shipped lockstep with the editor, and using the extension itself still requires a Copilot subscription. Treat it as a reading codebase, not a maintained dependency.

Information

  • Websitegithub.com
  • AuthorsMicrosoft
  • Published date2025/06/27

Categories

More Items

GitHub
AI Agent2026

Provides a lightweight Python harness that turns LLMs into working agents with tool-use, skills, persistent memory, permission controls and multi-agent coordination. Ships with a CLI/React TUI, 43+ built-in tools, a plugin/skill system and the ohmo personal-agent for chat gateways. Best for developers prototyping agent workflows and multi-agent experiments.

GitHub
AI Model2026

Runs the Bonsai family of quantized LLMs locally (including vision-capable 27B): provides scripts and demo UIs to run 1-bit and ternary Bonsai models on macOS (Metal), Linux/Windows (CUDA/Vulkan/ROCm), or CPU, with long context, tool-calling and an optional Open WebUI agent demo.

GitHub
AI Client2025

Turns Chromium into a local-first AI browser with an embedded assistant that can summarise pages, extract structured data, automate web tasks, and run scheduled agents. Built as an open-source Chromium fork with 53+ built-in browser tools, 40+ app integrations, and support for BYO AI keys or fully local models (Ollama / LM Studio).