Awesome Agent Protocols

This is a curated list of protocols and standards for AI agents, organized by the layer of the stack each one addresses: foundations, tool and context, agent to agent, agent to UI, agent to editor, serving and runtime, identity and discovery, the agentic web, payments and commerce, formats and conventions, governance, surveys and examples.

Each entry is marked as a wire protocol, a spec or standard, a format or convention, or deprecated or merged, and includes a short description of what it does. Entries include the Model Context Protocol with its tools, resources, prompts, sampling, elicitation and roots sub-specifications, the OpenAI Apps SDK, and editor-tooling precedents such as the Language Server Protocol and Debug Adapter Protocol that predate agent protocols but established the same client-provider decoupling pattern.

The list is released under CC0-1.0.