Back to directory

Recurse

A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

Recurse

Recurse is a serverless harness for creating and deploying specialist AI agents. A coding agent uses Recurse to design, test, and refine agent variations until it produces a specialist that meets a stated bar, then deploys it as a tool, an MCP server, or a bot.

Each agent is defined by a manifest file that pins its prompt, its input schema with validation and defaults, its output schema, and the specific tools it may call; Recurse refuses a manifest it does not recognize, and every successful run returns values matching the declared output schema. The coding agent owns the design loop, while Recurse's engine executes the resulting agent.

New accounts start with a small amount of free credit and no card is required to begin.

More in Agent frameworks

See all 902
  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.

  • A Windows UI Automation backend for computer-use agents that clicks, types and navigates real Windows applications through a fast, structured decision loop instead of a screenshot-and-guess approach, so a frontier model is not needed for every click.