Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: review the provided Terraform or Helm plan end-to-end for security, cost and drift, and return the full review including a ready-to-post PR comment and the monthly cost estimate.
- analyze — Run pure deterministic analysis over a parsed plan with no model call and return the findings, cost delta and drift count.
- scan_security — Return only the security findings — ports opened to 0.0.0.0/0 and publicly readable S3 buckets.
- estimate_cost — Return only the monthly cost summary — before, after and the delta in USD, computed from the figures in the plan.
- detect_drift — Return only the drift findings and the drift count — resources whose live state differs from the committed config.
- check_least_privilege — Return only the least-privilege findings — IAM policies granting wildcard actions or resources.
- list_capabilities — List the agent's static capabilities: the available tools and the security, cost and drift checks it runs. 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 infrastructure PR or the parsed plan and the review 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.
