Back to directory

Agentic Commerce Protocol (ACP)

An open standard, maintained by OpenAI and Stripe, that defines how buyers, their AI agents and businesses complete a purchase together using existing commerce infrastructure.

Agentic Commerce Protocol (ACP)

Agentic Commerce Protocol (ACP) is an open interaction model, currently in beta, for connecting buyers, their AI agents and businesses to complete purchases. Businesses make their products available for purchase through AI agents while continuing to use their existing commerce infrastructure and without becoming merchant of record for the agent's channel; AI agent developers can embed commerce so a user can discover and transact with a business directly inside an application; and payment providers process the resulting transactions by passing secure payment tokens between buyers and businesses through the agent.

The specification is maintained by OpenAI and Stripe and is versioned by date, with each version a complete snapshot: new work happens in an unreleased folder, is snapshotted to a dated release when ready, and older versions remain available but marked deprecated in the changelog. The repository publishes a human-readable RFC with design rationale and rollout plans, a machine-readable OpenAPI specification for checkout endpoints, JSON Schema data models, sample requests and responses, and a version changelog, and the protocol has reference implementations from both OpenAI and Stripe. It is licensed under Apache 2.0.

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.