#ai-tools

MCP server and Claude Code skill for Excalidraw — programmatic canvas toolkit to create, edit, and export diagrams via AI agents with real-time canvas sync.

Sessions, transactions, and version history for your knowledge graph. Operate without becoming the authority.

Analyze suspicious commits via code patterns, velocity anomalies, and statistical markers. Scan websites for AI generated text using pattern detection and optional OpenAI validation.

An agent composition framework that builds stacks and compiles specialized subagents for Claude Code

Switch between multiple Claude Code providers (Kimi, GLM, MiniMax, etc.) with a single command.

Google Gemini MCP - Use Google's visual, video and chat capabilities in your AI Assistant. Uses MCP Apps to preview created images, landing pages, SVGs and video

该仓库未发布描述。

Autonomous agent ecosystem — CLI tools that provision, scaffold, and maintain Claude Code agent ecosystems

Hot In AI (hotin) — what's hot in AI, from your terminal. Open-source CLI: GitHub + HN + npm + Reddit + "smart-money" (our own rubric ranked by corroboration & freshness).

Open-source interoperability layer for Coding Agents & LLMs

Local-first terminal app for learning Python, TypeScript, Java, and Rust with executable lessons, spaced review, coding practice, and optional AI hints.

Opinionated and idiosyncratic AI toolkit

Know how much Claude Code your friends are burning through. Create a group, share a code, compare usage.

The localhost AI Agent Runtime - Chat UI, Tools, RAG, and MCP in one pip install

该仓库未发布描述。

⚡ The AI Slop Index linter inspired by @karpathy. Detects hallucinated imports, any-type abuse, vibe coding, and soul-less patterns in TypeScript/JavaScript/React. Fast (~4s), educational, CI-ready. npx karpeslop@latest

Warm-process MCP server for PHPUnit — ~4× faster per test call. Works with Claude Desktop, Cline, Continue, Cursor, Zed, any MCP client.

Select text in Claude Code, Codex CLI, GitHub Copilot CLI or Kiro CLI output and get a short explanation right under it, inside the terminal. Or ask your own question about it.

Verify that every PR review comment was addressed. CLI + GitHub Action for agentic and human review workflows.

Helper library for C# Projects

Switch Claude Code between LLM providers — no proxy, no config editing