Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: generate the full ad copy variant set across the chosen platforms, weighted by the past message-axis performance you supply, run the pre-publish policy scan and return the variants with the findings.
- analyze — Run pure deterministic message-axis performance analysis over your past CTR data with no model call, returning scored findings with their sample sizes and confidence.
- generate_variants — Generate the variant set on its own — headline and description pairs per platform, weighted toward winning axes and truncated to each platform's character limits.
- scan_policy — Run the pre-publish policy scan and return only the variants that tripped a filter — superlative, guarantee or medical-claim phrasing.
- check_platform_limits — Check each variant's headline and description length against the target platform's limits and report the fit per variant.
- list_capabilities — List the agent's static capabilities: the available tools, the supported platforms and message axes, and the CTR benchmark used to rank findings. 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 product or offer, the platforms and any past CTR you have. 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.
