Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: apply the attribution model to your channel data, compute the efficiency metrics per channel and return the full report and dashboard with the budget recommendation.
- analyze — Run pure deterministic attribution and efficiency analysis over the provided channel data with no model call, returning scored per-channel findings.
- apply_attribution — Apply the multi-touch attribution model and return the attributed revenue per channel — the redistribution step alone, useful for comparing two models before any metric is computed.
- compute_channel_metrics — Compute the full per-channel table: spend, attributed revenue, CAC, LTV, LTV to CAC, payback period, ROI and the budget recommendation.
- rank_findings — Return the scored per-channel findings sorted worst-first, from critical through warning to healthy.
- recommend_budget — Return the per-channel budget shift — increase, maintain or decrease — with the signals behind each call.
- list_capabilities — List the agent's static capabilities: the available tools, the supported attribution models and the efficiency 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 reporting period, your per-channel numbers and the attribution model. 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.
