AgentField

What is AgentField?

AgentField is an open-source control plane for running AI agents as production services. You write agent logic as ordinary functions in Python, Go or TypeScript, and AgentField exposes each one as a REST endpoint so frontends, backends, cron jobs and other agents can call it like any other API. The control plane is a stateless Go service that handles routing, async execution, retries, webhooks, memory and observability, and it can fan a single request out to thousands of agents. Governance features give each agent a W3C DID identity, verifiable credentials for each execution and tag-based access policies. It also includes pause-for-approval steps, canary and blue-green rollouts of agent versions, and a way to dispatch tasks to coding agents such as Claude Code or Codex. The code is open source under Apache-2.0, so there is no license fee, though you still pay for the model APIs and infrastructure you connect.

Use cases

• Expose AI agents as REST endpoints
• Fan one request out to many agents
• Trace and retry multi-agent workflows
• Pause agent runs for human approval
• Roll out agent versions with canary traffic
• Run coding agents under orchestration

Who it's for

• Backend engineers
• Platform teams
• AI engineers
• DevOps teams

Why use AgentField

• Agents become callable services, so any other part of your stack can invoke them without custom glue code.
• Queues, retries, backpressure and traces are handled by the control plane, which removes work you would otherwise write by hand.
• Cryptographic identities and signed audit records give agents more accountability than a shared API key.
• You keep the agent framework you already use, since a reasoner is a plain function that can wrap LangGraph or CrewAI code.

AgentField features

• Every function becomes a REST endpoint
• Python, Go and TypeScript SDKs
• Async execution with webhooks and SSE streaming
• Cross-agent calls routed with full tracing
• Agent discovery by tag across the mesh
• Memory with key-value and vector search in four scopes
• app.pause() for durable human approval
• Canary, A/B and blue-green agent rollouts
• W3C DID identity and verifiable credentials per agent
• Tag-based policy enforcement
• Workflow DAGs and Prometheus metrics

AgentField integrations

• LangGraph
• CrewAI
• Claude Code
• Codex
• Gemini CLI
• OpenCode
• Cursor
• Windsurf
• Aider
• Docker
• Kubernetes
• Prometheus
• LiteLLM

How to get started with AgentField

1. Install the af CLI with the install script, or run the control plane with docker run -p 8080:8080 agentfield/control-plane:latest.
2. Scaffold an agent with af init my-agent, choosing Python, Go or TypeScript.
3. Write reasoner or skill functions and call app.run(), which registers the agent and exposes each function as an endpoint.
4. Call it with a POST to /api/v1/execute/<agent>.<function>, or use the async and streaming endpoints.
5. Watch workflow DAGs and execution traces in the dashboard at localhost:8080.

AgentField limitations

• Routing adds roughly 100-200ms per cross-agent hop
• You run and operate the control plane yourself, including Docker or Kubernetes setup
• Model API costs and infrastructure are separate from the open-source code

AgentField pricing

AgentField is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

AgentField alternatives

The main AgentField alternatives listed on Debuly are LangGraph, CrewAI, Mastra, Agno, AG2 and Microsoft Agent Framework. The table compares pricing and platforms; order is A–Z.

1. 

   AG2

   Open-source multi-agent framework, formerly AutoGen, plus an AgentOS platform for coordinating and deploying teams of AI agents.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit AG2 (official site)
2. 

   Agno

   Python framework and AgentOS runtime for building agents, teams and workflows, then serving them as an API with a control plane.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Agno (official site)
3. 

   CrewAI

   Multi-agent framework and platform for building AI agent crews in Python or a visual editor, then deploying and monitoring them.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit CrewAI (official site)
4. 

   LangGraph

   Low-level open-source orchestration framework and runtime for building stateful, long-running AI agents as graphs.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit LangGraph (official site)
5. 

   Mastra

   Open source TypeScript framework for building AI agents and workflows, plus a hosted platform for Studio, Server, memory and observability.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Mastra (official site)
6. 

   Microsoft Agent Framework

   Open, multi-language framework from Microsoft for building, orchestrating and running AI agents and multi-agent workflows in Python, .NET and Go.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Microsoft Agent Framework (official site)

AgentField FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from avatars.githubusercontent.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.