Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: from a provided security-audit report, produce one single-package upgrade PR proposal per vulnerable package — version bump, breaking-change detection, migration steps and test result — plus a risk summary.
- analyze — Run pure deterministic analysis over a provided audit report with no model call and return one single-package PR proposal per vulnerability.
- classify_semver_jumps — Classify the semver jump — patch, minor, major or none — for each vulnerable package in the audit.
- score_upgrade_risk — Score the breaking-change and upgrade risk for each vulnerable package, returning a deterministic risk score and band.
- plan_migrations — Return the deterministic, ordered migration steps for each package's upgrade.
- list_upgrade_prs — List the single-package upgrade PR proposals — title, version bump, merge autonomy and test result, one per package.
- summarize_upgrades — Return the headline counts across all proposed upgrades: vulnerable packages, PRs, high and critical CVEs, major jumps and auto-merge candidates.
- list_findings — Return the per-package severity findings with label, risk band and detail.
- list_capabilities — List the agent's static capabilities: the available tools and a short description of the upgrade pipeline. 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 repository, ecosystem and audit report. 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.
