Farna Control
Read-only demoInstall

Feedback System

A rejection carries information a bare failure does not. When a reviewer rejects a pipeline output at a human-in-the-loop (HITL) gate, the Feedback System records why, decides what should happen next, and can hand the run to a correction pass that keeps the original context.

Core concepts

  • Feedback Record: Created when a HITL reviewer rejects a pipeline output. Stores rejection reason, rejected output, producing node/agent, and routing info.
  • Feedback Handler: Determines how a rejection is triaged: manual review (human), automatic AI correction (ai_correction), or AI correction with human review (ai_correction_with_human_review).
  • Correction Run: A new LangGraph run spawned from a feedback record, pre-configured with the original run’s context and the rejection details.
  • Feedback Inbox: UI surface for reviewing, annotating, and resolving feedback records across all pipelines.

API endpoints

API endpoints
Endpoint Method Description
/api/v1/runs/{run_id}/feedback POST Create a feedback record for a run
/api/v1/feedback GET List feedback records (paginated)
/api/v1/feedback/{record_id} GET Get a single feedback record
/api/v1/feedback/{record_id}/status PATCH Set the record’s status: pending, routing, correcting, resolved, escalated, or dismissed
/api/v1/feedback/{record_id}/detect-gap POST Run the pipeline’s evals against the rejected output and flag an eval gap when none catches it
/api/v1/feedback/inbox GET List the feedback inbox
/api/v1/feedback/inbox/{record_id} GET Get an inbox entry, including the pipeline name
/api/v1/feedback/inbox/{record_id}/review POST Review action: mark_reviewed, dismiss, or create_correction_run, with an optional annotation
/api/v1/feedback/proposals GET List eval-gap proposals (paginated)
/api/v1/feedback/proposals/{record_id}/publish POST Publish an eval-gap proposal as a live eval definition

Routing strategy

A pipeline-level default_feedback_handler applies to all HITL gates unless a gate overrides it with its own feedback_handler. When both reject_target and feedback_handler are set on the same gate, feedback_handler takes precedence.

Eval gap detection

When a rejection is recorded, the system can retroactively run the pipeline’s eval suite against the rejected output. If no existing eval catches the failure, the record is tagged as an eval_gap, indicating an opportunity to expand the eval suite.

Next: Triggers