kitaru

About kitaru

Traces you can run, not just read. Kitaru (来る, "to arrive") records every agent run as a full trace — every model call, tool call, and decision — and replays it against your real code. Reproduce the trace exactly. Fork it with one thing changed. Trust the diff. It works underneath whatever framework you already use, self-hosted on your own infrastructure, and it can deploy and run your agents too. Docs · Quick Start · Examples · Getting Started Guide · Roadmap · Community

Quick Facts

Compatible Skills

These tools work well together with kitaru for enhanced workflows:

• agentcanvas — semantic(0.43)+complementary+rare_topics+same_lang+similar_pop+shared_platform (69%)
• pydantic-ai-todo — semantic(0.41)+complementary+rare_topics+same_lang+similar_pop+shared_platform (68%)
• pydantic-ai-shields — semantic(0.35)+complementary+rare_topics+same_lang+similar_pop+shared_platform (66%)
• pydantic-ai-skills — semantic(0.33)+complementary+rare_topics+same_lang+similar_pop+shared_platform (61%)

More MCP Server Tools

Explore other popular mcp server tools:

• rea ⭐ 335.5k
• ECC ⭐ 283.7k
• n8n ⭐ 208.0k
• everything-claude-code ⭐ 186.4k
• dify ⭐ 158.9k
• open-webui ⭐ 156.2k
• cc-switch ⭐ 146.2k
• graphify ⭐ 127.2k
• browser-use ⭐ 118.3k
• gemini-cli ⭐ 107.5k

Popular Python Agent Tools

• hermes-agent ⭐ 255.2k · Codex Skill
• skills ⭐ 179.5k · Agent Tool
• dify ⭐ 158.9k · MCP Server
• open-webui ⭐ 156.2k · MCP Server
• langflow ⭐ 155.8k · Agent Tool

Frequently Asked Questions