Your coding agent uses Recurse to define Python tools and checks around a task. An LLM calls those tools to explore alternatives and refine results. Execution is serverless: expose the workload as an MCP server for use in agentic workflows, or run it to completion. Requires the Recurse SDK and a Recurse account.
Install
- 1Add it in ChatGPT
Open Plugins in the ChatGPT sidebar (web or desktop), search for “Recurse” and select +. You'll be asked to sign in to any connected service on first use. Then type @ in a chat to call it.
Skills 1
- recurse
Build and run custom agents with Recurse for tasks that benefit from test-time compute and iterative refinement. Examples include design exploration, performance optimization, and generating artifacts that improve through verification and revision. Use this skill when designing an adaptive component that solves a specific problem. When necessary, clarify uncertain requirements on the goal and constraints.
Things to ask
- “Design a Python tool that tests code changes against a benchmark.”
- “Plan a Recurse agent that generates puzzle levels and checks solvability.”
- “Help me expose a tested Recurse agent through MCP.”
Related
SuperpowersPlan, develop, and debug codeplugin · by Jesse Vincent · 294k stars
ShopifyBuild and manage your storeplugin · by Shopify · 7.2k stars