Extensions

Blink - Local AI Code Completions

Lojell

Lightning-fast inline AI completions (ghost text), local-first and BYOK: GGUF models in-process via llama.cpp, or any OpenAI-compatible endpoint.

Claude Gate

lntvan166

Review Claude Code's file changes in a native diff panel — accept or reject each edit. The PreToolUse hook captures all Claude Code edits (terminal CLI and in-editor extension); the filesystem watcher is an optional fallback for non-Claude agents and is off by default.

Agentic HTML Visual Editor

YuyaMiyamoto

A WYSIWYG HTML editor for reviewing and refining AI-generated content in VS Code

AI Diff Viewer

mtglitch

Calm review of AI CLI edits — inline diff hunks in your normal editor, opt-in split-view diff, hover actions. For Claude Code, Codex, and Qwen.

RuntimeAds

runtimeads

Get paid while Claude Code and Codex think. Privacy-safe sponsor ads during AI wait time.

Provenance

etpedror

Embed requirement and compliance traceability directly in your code. Keeps AI assistants from removing load-bearing logic they don't understand.

Markr – AI Config Editor & Markdown Preview

Apptware-Product-Lab

The memory & control layer for AI coding tools + Markdown preview — Context Bridge hands off sessions between Claude Code, Cursor, Augment, Codex, Cline, Windsurf & Gemini; cross-session memory, self-writing configs, AI Scoreboard; AI Agent Map for multi-agent .claude/ repos, Config Lab to test your

Opencoder — A Clean Opencode Interface

siilver

A clean and polished VS Code interface for OpenCode

Markify

MCCProgrammer

Save tokens with markdown

Skill Curator · 技能管家 (多模型 AI 技能管理)

Xianwei-Zhang

Xiaomi MiMo（小米）、DeepSeek（深度求索）、Qwen（通义千问）、Gemini、OpenAI 多模型 AI 技能管理器：扫描并用大模型诊断、同步本机 Claude Code / Codex / Copilot / Gemini / OpenCode 的 skills（技能）。Manage, LLM-diagnose & sync your AI coding-agent skills across multiple model providers.

Cloud LM Provider — AWS Bedrock & Azure OpenAIfor GitHub Copilot

suddhu-iith2004

🚀 The #1 extension for enterprise AI in VS Code! Bring Claude 4.5, GPT-4o, Nova, DeepSeek & 50+ models into Copilot Chat. Features Headroom AI compression that saves 30-45% on API costs. Secure, fast, with full streaming, tool calling & vision support.

AI-first project governance — context, policies, and prompt optimization for vibe coding