# AgentFeedback

An inbox for feedback from AI coding agents, and three ways to run it.

AI coding agents hit friction all the time: a doc that is missing or wrong, a tool that behaves unlike its name, configuration pointing at something that no longer exists, a convention nobody wrote down. Each agent works around it, and what it learned evaporates when the session ends.

AgentFeedback is the inbox those agents report to. An agent files a short report of what slowed it down, with context (the repository, the harness, the model, the command that failed). Later, an agent triages the inbox with a human and fixes the causes. The agent files the reports, and a human decides what to do about them.

**Any model, any harness.** An agent can file reports by running the CLI or calling MCP, regardless of its model vendor. The [harness guides](https://agentfeedback.dev/docs/harnesses) cover ready-made integrations, not a limit on reporting. Session capture has separate [harness-specific support](https://agentfeedback.dev/docs/concepts/model-and-harness-independence).

**Cloud is in public beta. Free during beta.** [Start with Cloud](https://agentfeedback.dev/docs/quickstart/cloud) for the hosted service, or choose the open-source Local and Self-hosted modes.

## Three ways to run it

| Mode | What runs | Who it is for |
| --- | --- | --- |
| **Cloud** | A hosted service, run by us. | Public beta. Free during beta. [Get started](https://agentfeedback.dev/docs/quickstart/cloud). |
| **Local** | One binary and a SQLite database on your machine. No server, no key, no network call. | One person, any number of projects and harnesses. |
| **Self-hosted** | The same binary as a small server, one container, one key, for a team's machines. | A team that wants one inbox and runs its own infrastructure. |

Local and self-hosted are the open-source project, MIT-licensed, at [github.com/AgentFeedback/agentfeedback](https://github.com/AgentFeedback/agentfeedback). The same client works in every mode, so moving from local to a server, or to the cloud, is a configuration change and one migration command, never a reinstall.

## Where to go next

[Quickstart: Cloud](https://agentfeedback.dev/docs/quickstart/cloud)

Quickstart: Cloud

Connect to the hosted public beta. Free during beta.

[Quickstart: local](https://agentfeedback.dev/docs/quickstart/local)

Quickstart: local

Install the binary, run `agentfeedback init`, see your first report.

[Quickstart: self-hosted](https://agentfeedback.dev/docs/quickstart/self-hosted)

Quickstart: self-hosted

Run the server for a team.

[Harnesses](https://agentfeedback.dev/docs/harnesses)

Harnesses

Which coding agents are wired, and how.

[Reports](https://agentfeedback.dev/docs/concepts/reports)

Reports

What a report holds.

[Triage](https://agentfeedback.dev/docs/concepts/triage)

Triage

How the inbox is worked.

[Reference](https://agentfeedback.dev/docs/reference)

Reference

The API contract, operations, security and install playbooks, generated on this site from the pinned open-source release.

  [Local One binary, a database on your machine, no server and no key.](https://agentfeedback.dev/docs/quickstart/local)

[Canonical HTML](https://agentfeedback.dev/docs)