Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: run the week-over-week analysis over the competitor watch-list you provide and return the full weekly competitive report.
- analyze — Run pure deterministic analysis over the provided watch-list with no model call and return the scored findings plus the structured weekly report.
- list_new_ad_creatives — Return the new ad creatives discovered per competitor, with platform and how recently each was first seen.
- detect_messaging_shifts — Return the messaging-shift analysis per competitor — headline word-overlap against their prior baseline, with the shift ratio.
- diff_pricing — Return the pricing-page diffs per competitor: the plan, the price before and after, and the delta.
- track_serp_moves — Return the won and lost search position moves per competitor keyword, with the from and to positions.
- list_coverage_gaps — Return the ad-library coverage gaps — which competitors and countries had partial or unavailable coverage — so an incomplete picture is never presented as a complete one.
- list_content_velocity — Return each competitor's publishing velocity inside the lookback window, with the count and the titles.
- list_findings — Return the scored findings alone, worst first.
- 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 competitors, the country and the lookback window. 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.
