Extensions

MCP Server Authentication

VilashJagani

Multi-method authentication for MCP servers: Device Auth Flow, Client Credentials, Password Grant, Basic Auth, and API Key — per-server config, per-IDP settings, injected into VS Code MCP config automatically.

AI #ai

99 5.0

DeepSeek Family for GitHub Copilot Chat

Dootmaan

Bring DeepSeek V4-Pro and V4-Flash (1M context, selectable thinking effort) into GitHub Copilot Chat with visible thinking.

Programming Languages AI #ai

98 5.0

Rovo Tools for Copilot

manulife-internal

Atlassian Rovo semantic search and fetch as native VS Code language model tools for GitHub Copilot agent mode.

AI #language-model-tools

97 5.0

Antigravity CLI Copilot

jacobthejacobs

Gemini chat in a VS Code side panel, powered by Google's Antigravity CLI (agy). Streaming replies, tool calls, inline images. Unofficial.

Machine Learning AI #ai

96

Selenium Boot MCP — AI Browser Automation

seleniumboot

MCP server for Selenium WebDriver automation. Lets GitHub Copilot and Claude control a real browser — 85 tools for navigation, interaction, assertions, and test codegen.

Testing AI #testing

95

ByNara AI

byNara-id

VS Code extension for the ByNara AI coding agent

AI Chat #keybindings

95

MCP Server for VS Code

nabheet

MCP server that lets AI agents (opencode, Claude, Cursor) debug, edit, and control VS Code directly. Debugger, terminal, LSP, file ops.

Programming Languages Debuggers #remote

91

GitHub Copilot Model Advisor

cperezsx

A VS Code sidebar advisor that ranks enabled GitHub Copilot models by health, latency, and task fit.

AI #ai

90 5.0

Pi / Code

auchan

A Pi-native multi-session coding agent workspace for VS Code. Pi Coding Agent / Pi on Code / Pi Code GUI.

AI Chat #ai

89 5.0

Copilot & BYOK Usage Tracker

feima

BYOK usage tracking, cost dashboard, and session analysis for GitHub Copilot & BYOK (DeepSeek, Mistral, OpenRouter, Qwen, Kimi K3, GLM/Z.ai, MiniMax, MiMo). Manage providers visually; browse 100+ AI coding tools.

Visualization AI #ai

89

Claude Savings

Decompute

Cut your Claude Code token bill. A local gateway compresses context before each request — same model, same answers, fewer tokens. Bring your own key; runs entirely on your machine.

Machine Learning AI #claude code

89 5.0

DevPass (LLM Gateway)

grosch-capital

Adds DevPass by LLM Gateway as a language model provider in VS Code Chat. Pre-configured endpoint; just add your API key and models are discovered automatically.

AI Chat #ai

88