autoresearch-claude-code

About autoresearch-claude-code

autoresearch-claude-code A measured experiment loop for Claude Code and Codex. Define a goal, a fixed benchmark, and files to modify. The agent tries changes, measures results, commits supported improvements, and records failed ideas within a budget. Inspired by pi-autoresearch. The experiment protocol is a shared skill with standard-library Python helpers; no MCP server is required. Install Requires Python 3.10+, Bash, Git, and your chosen agent client. Linux and macOS are supported; use WSL on Windows. The optional ML example has its own dependencies.

Quick Facts

Compatible Skills

These tools work well together with autoresearch-claude-code for enhanced workflows:

• ResearcherSkill — semantic(0.28)+complementary+same_lang+similar_pop+shared_platform (55%)
• evo — semantic(0.38)+complementary+same_lang+similar_pop (53%)
• Agent-Loop-Skills — semantic(0.24)+complementary+same_lang+similar_pop+shared_platform (53%)
• Thoth — semantic(0.35)+complementary+same_lang+similar_pop (52%)
• social-account-doctor — semantic(0.19)+complementary+same_lang+similar_pop+shared_platform (52%)

More Claude Skill Tools

Explore other popular claude skill tools:

• claude-mem ⭐ 101.8k
• Understand-Anything ⭐ 86.0k
• archify ⭐ 85.8k
• codegraph ⭐ 73.7k
• oh-my-openagent ⭐ 70.2k
• get-shit-done ⭐ 63.8k
• diagram-design ⭐ 58.3k
• marketingskills ⭐ 55.5k
• Understand-Anything ⭐ 53.4k
• scientific-agent-skills ⭐ 52.8k

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