Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: analyse the provided infra and APM signals into scored findings, compute the before/after benchmark, draft the PR suggestion and return the full performance report.
- analyze — Run pure deterministic analysis over provided infra and APM data with no model call and return scored performance findings across the database, loop, duplication, cold-start and latency categories.
- benchmark — Compute the deterministic before/after benchmark on its own — p50 and p95 latency, throughput, error rate and improvement percentage — derived from the analysed findings and your baseline sample.
- list_capabilities — List the agent's static capabilities: the available tools and the analyzer's warning and critical thresholds for each metric. 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 target, its type and the run 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.
