Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: compare the provided acceptance criteria against what shipped and return the full gap report — findings, coverage counts and follow-up ticket suggestions.
- analyze — Run pure deterministic analysis over the provided comparison data with no model call and return scored gap findings.
- list_gaps — Return only the gaps — every acceptance criterion that is not fully met, whether partial, unmet or silently dropped, scored worst-first.
- list_unmet — Return only the acceptance criteria that were not delivered at all.
- list_silently_dropped — Return only the criteria that were silently dropped from scope with no PRD amendment — the highest-surprise gaps in a release review.
- suggest_tickets — Return one follow-up ticket suggestion per criterion that is not fully met, with a title, a rationale and a priority.
- summarize_coverage — Return the headline coverage counts: total acceptance criteria broken down by met, partial, unmet and silently dropped.
- list_capabilities — List the agent's static capabilities: the available tools and the mapping from criterion status to severity. Useful for discovery.
- plan_inputs — Plan the inputs for a tool: returns the clarifying questions, the JSON schema and a ready-to-edit example.
- discover_intent — Understand your goal and co-design the exact input through clarifying questions before running.
- open_form — Opens the guided form to collect the PRD or feature key and the acceptance criteria. The entry point for running the agent on your own data.
- run_form — Internal: invoked by the guided form when the user submits. Not for direct use.
