task-completition-evaluation

Task completion validation aggregates diffs, CI tests, and collaboration evidence to enforce PR readiness. It compares implementation against original requirements to ensure cross-agent communication, documented handoffs, and passing checks before declaring a task complete.

A task completion evaluation gate is a final audit step that verifies multi-agent communication and documented handoffs. It prevents silent or partial completions by collecting GitHub PR health data and Slack eval transcripts to support a definitive completion verdict.

PR readiness enforcement consolidates GitHub PR health data and Slack eval transcripts to verify CI checks pass and reviews are complete. This evidence aggregation ensures necessary docs are updated and cross-agent communication is documented before final closure.

Diff review and CI tests are aggregated to prevent partial task handoffs by verifying all artifacts are present. The evaluation gate checks multi-agent communication, GitHub PR status, and documented evidence to ensure no silent completions occur.

Running a final completion check involves triggering an end-to-end evaluation flow after implementation finishes. This gate produces a final verdict by validating diff reviews, cleanup, real tests, and PR readiness against the original task requirements.

A final evaluation gate is needed at the finish of implementation to ensure PR health, cross-agent communication, and documented evidence. It enforces requirements like diff review, cleanup, real tests, and PR readiness before declaring task completion.