# Agent Feedback

## Docs

- [Reading Agent Feedback from HTTP and HTML Responses](https://docs.epode.ai/agents/http-html.md): Learn the three places Agent Feedback embeds contracts in HTTP/HTML responses and how to parse them with the Node, Python, Go, and Rust agent helpers.
- [Submitting Outcome Reviews from Feedback-Aware Agents](https://docs.epode.ai/agents/submitting-outcomes.md): Submit success, partial, or failure outcome reviews from your agent runtime using the Agent Feedback SDK helpers for Node, Python, Go, and Rust.
- [Agent Feedback Well-Known Service Discovery Endpoint](https://docs.epode.ai/api/discovery.md): Resolve Agent Feedback service capabilities, SDK package download URLs, and submission endpoints via the versioned well-known discovery document.
- [Submit Agent Outcome Review — POST /api/v2/outcomes](https://docs.epode.ai/api/outcomes.md): Submit a compact success, partial, or failure outcome review from a customer agent using a scoped afr2_ capability token. Idempotent and privacy-enforced.
- [Agent Feedback API Reference and Authentication Guide](https://docs.epode.ai/api/overview.md): Base URL, product key and capability token authentication, rate limits, response format, and error codes for the Agent Feedback REST API.
- [Send Telemetry Batches — POST /api/v2/telemetry/batches](https://docs.epode.ai/api/telemetry.md): Reference for the SDK telemetry batch endpoint. Ships interaction opportunities asynchronously with your af_live_ product key without blocking responses.
- [Capability Tokens: Scoped Agent Credentials (afr2_…)](https://docs.epode.ai/concepts/capability-tokens.md): Learn how Agent Feedback capability tokens work, what claims they carry, how the SDK signs them locally, and what the backend verifies on submission.
- [Agent Feedback Dashboard: Products, Keys, and Teams](https://docs.epode.ai/concepts/dashboard.md): Navigate the Agent Feedback dashboard to create products, manage API keys, review outcomes and sessions, set collection policy, and manage your team.
- [Feedback Contract: The _agentFeedback Envelope](https://docs.epode.ai/concepts/feedback-contract.md): Understand the _agentFeedback envelope Agent Feedback embeds in eligible responses, how it reaches agents across JSON, HTML, and header surfaces.
- [Interactions: How Agent Feedback Tracks Responses](https://docs.epode.ai/concepts/interactions.md): Learn how Agent Feedback defines interactions, how they are classified, and what gets promoted from an unclassified opportunity to a confirmed review.
- [Collection Policy: Control Which Responses Get Feedback](https://docs.epode.ai/configuration/collection-policy.md): Learn how the dashboard collection policy and SDK-level route filters work together to control where feedback contracts appear in your product responses.
- [Configuration Options Reference for Agent Feedback SDK](https://docs.epode.ai/configuration/options.md): Complete reference for all AgentFeedbackOptions fields: API key, endpoint, route filters, feedback mode, context extractors, and queue tuning.
- [How Agent Feedback Works: The Full Protocol Flow](https://docs.epode.ai/how-it-works.md): A technical walkthrough of the Agent Feedback protocol — from capability token creation and envelope embedding to outcome submission and telemetry.
- [Agent Feedback: Outcome Signals for AI-Driven Products](https://docs.epode.ai/introduction.md): Learn how Agent Feedback collects compact, privacy-first outcome signals from the autonomous AI agents using your product — with zero added latency.
- [Quickstart: Add Agent Feedback to Your API or Service](https://docs.epode.ai/quickstart.md): Instrument your Node/Express app to collect AI agent outcome signals in five steps — with Python, Go, and Rust examples included.
- [Go SDK: net/http Middleware for Agent Feedback Apps](https://docs.epode.ai/sdks/go.md): Wrap any net/http handler with Agent Feedback using zero external dependencies. JSON and HTML responses are instrumented; streams are left untouched.
- [MCP Server Instrumentation with the Agent Feedback SDK](https://docs.epode.ai/sdks/mcp.md): Instrument any MCP server with one function call. Tool calls become confirmed agent interactions with an explicit protocol tool for submitting outcomes.
- [Node.js SDK: Express, Fastify, and MCP Instrumentation](https://docs.epode.ai/sdks/node.md): Instrument your Express or Fastify app with Agent Feedback. One middleware call adds outcome feedback to every eligible JSON, HTML, or header response.
- [Python SDK: ASGI and WSGI Middleware for Agent Feedback](https://docs.epode.ai/sdks/python.md): Add Agent Feedback to any Python web app with one middleware line. Supports FastAPI, Starlette, Flask, Django, and every other ASGI or WSGI framework.
- [Rust SDK: Agent Feedback Tower Layer for Axum Apps](https://docs.epode.ai/sdks/rust.md): Add Agent Feedback to any Axum application with a Tower layer. Finite JSON and HTML bodies are instrumented; bounded streams are left untouched.
- [Integration Doctor: Verify Your Feedback Contract End-to-End](https://docs.epode.ai/tools/integration-doctor.md): Run agent-feedback-doctor against a live URL to verify the feedback contract is correctly embedded and a synthetic outcome review is accepted end-to-end.
- [Agent Feedback Protocol v1: SDK and Conformance Reference](https://docs.epode.ai/tools/protocol.md): Language-neutral Agent Feedback Protocol v1: signing algorithm, envelope shape, telemetry batch format, and the conformance test vector.