# agi-inc/agent-protocol

Common interface for interacting with AI agents. The protocol is tech stack agnostic - you can use it with any framework for building agents.

Repository: https://github.com/agi-inc/agent-protocol
Canonical: https://ross.abutalabs.com/products/agent-protocol
Homepage: https://agentprotocol.ai
Language: Python
License: MIT
License Family: permissive
Topics: agents, ai, api, javascript, llms, openapi, protocol, python, typescript, ai-agent, auto-gpt, gpt-4
Last push: 2025-04-08T06:10:21+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 15, release rhythm 8, longevity 82
- inputs: {"age_days": 1154, "days_push": 512, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1455, forks 186 (observed 2026-08-28T04:04:46.830009+00:00)

## What it is
Agent Protocol is an OpenAPI-defined REST API specification providing a common interface (runs, tasks, steps, artifacts) for interacting with AI agents, regardless of the framework used to build them. It ships SDKs (Python, TypeScript) so agent developers can adopt the standard without writing boilerplate API code.

## Use cases
- standardize the API interface for my AI agent
- make my agent compatible with benchmarks and devtools
- avoid writing boilerplate REST endpoints for an agent
- integrate with agents built in different frameworks
- compare and evaluate multiple agents through one interface
- expose my agent as a REST service with tasks and artifacts

## When to choose
- you want a framework-agnostic REST API standard for driving agents
- you need agents to be benchmarkable or integrable via a common interface
- you want Python or TypeScript SDKs to implement the spec quickly

## When to avoid
- you need agent-to-tool context sharing (use MCP instead)
- you need agent-to-agent task delegation (use A2A instead)
- you need a rapidly evolving standard with broad current industry adoption

## Facets
- artifact type: library
- maturity: maintenance
- function: api-framework, agent-framework, sdk, middleware
- domain: large-language-models, apis, developer-tools
- platform: python, cross-platform
- tags: openapi, rest-api, agent-protocol, interoperability, benchmarking-agents, typescript-sdk, framework-agnostic, ai-agents, nodejs

## Member repositories
- agi-inc/agent-protocol (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:46.830009+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:35:38.246166+00:00, confidence not recorded.
  - readme: https://github.com/agi-inc/agent-protocol (fetched 2026-08-28T04:04:46.830009+00:00, sha a7696359f0ee)
  - homepage: https://agentprotocol.ai (fetched 2026-08-29T11:44:44.131641+00:00, sha 16de17aaf59a)
  - site_page: https://agentprotocol.ai/about (fetched 2026-08-29T11:44:44.142678+00:00, sha 7705934a131f)
  - site_page: https://agentprotocol.ai/faqs (fetched 2026-08-29T11:44:44.140837+00:00, sha 07b988e9cea7)
- Data as of 2026-08-30T08:39:29.467469+00:00.