Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: design the segment-based drip series end-to-end — trigger, cadence, email drafts with A/B subjects, exit criteria and expected conversion impact. Always returned as a draft; send authority is never granted.
- analyze — Run pure deterministic guardrail analysis over the provided segment, goal and engagement baseline with no model call, returning scored findings.
- build_trigger_condition — Return only the entry condition that puts someone into the flow, derived from your segment.
- plan_cadence — Return only the send cadence — per step a day offset, an hour offset and a suggested send window in the recipient's local time — shaped by the flow family rather than the step count.
- draft_emails — Return only the per-step email drafts with A/B subject-line variants and body previews. Drafts — never sent.
- build_exit_criteria — Return only the exit criteria: when a contact leaves the flow, including routing an unconverted contact to human review rather than another automated send.
- estimate_impact — Return only the expected conversion impact — baseline against projected — from the engagement baseline you supply. A self-contradictory funnel is refused rather than projected.
- list_capabilities — List the agent's static capabilities: the available tools and the analyzer's 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 segment, the campaign goal and the number of steps. 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.
