Back to directory

Open Agent Specification

A platform-agnostic declarative language for describing agents and structured agentic workflows, defining the components, properties, and semantics needed to compose multi-agent systems.

Open Agent Specification

Agent Spec is a portable, platform-agnostic configuration language for describing AI agents and agentic systems with enough detail to run them. It defines the conceptual objects and components that make up typical agent systems, along with the properties that configure them and their semantics.

Two runnable component types anchor the language: agents, such as ReAct-style conversational agents, and flows, such as structured, workflow-based business processes. Runtimes implement these components for execution against specific agentic frameworks or libraries, and SDKs, starting with Python, serialize and deserialize agents and flows to and from JSON or YAML while checking conformance to the specification.

The Python package is distributed under the Apache-2.0 license.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    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.

  • 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.