Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: pool the feedback you provide, theme it, weight each theme by volume, ARR and segment, and return the full theme report and dashboard.
- analyze — Run pure deterministic theming and revenue weighting over a provided feedback pool with no model call, returning the scored themes.
- classify_feedback — Classify each feedback item into a deterministic theme label with no model call. Returns one row per item with its id, source and theme.
- rank_themes — Rank themes by revenue-weighted priority — 40% volume, 60% affected ARR — returning the priority score, severity and ticket volume per theme.
- revenue_impact — Weight each theme by revenue impact: affected ARR over DISTINCT accounts, plus ticket volume and the segments involved.
- theme_quotes — Surface representative quotes per theme, highest-ARR account first, sanitised.
- theme_trends — Show each theme's volume trend — up, down or flat — with the per-source breakdown behind it.
- recommend_backlog — Recommend a backlog item per theme, ordered by revenue-weighted priority. Proposals only — nothing is written to a tracker.
- list_capabilities — List the agent's static capabilities: the available tools, the supported sources and the theme taxonomy. 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 analysis window, the sources to include and how many themes to return. The entry point for running the agent on your own feedback.
- run_form — Internal: invoked by the guided form when the user submits. Not for direct use.
