Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: run the impact analysis end-to-end for a PR or commit and return the findings, risk matrix, breaking points, suggested regression tests and a narrative summary.
- analyze — Run pure deterministic analysis over provided code-graph data with no model call and return scored findings, one per affected module.
- build_risk_matrix — Project the risk matrix — affected module by severity, breaking flag and consumer fan-in, worst first.
- list_breaking_points — Return only the breaking call chains and contracts a reviewer must check before merging.
- suggest_regression_tests — Return the suggested regression test set — one per module with consumers, an unknown fan-in, a touched contract or a breaking change.
- score_pr_risk — Roll the affected modules into one merge decision: a 0-100 PR risk score, its band, the critical, warning and breaking counts, the consumer call sites at risk, and a merge recommendation.
- list_impacted_consumers — Name the consumer services that must be notified about this PR, and which affected modules reach them.
- list_capabilities — List the agent's static capabilities: the available tools and the analyzer's fan-in and breaking-change severity thresholds. 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 PR or commit and the scope. 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.
