Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: dedupe and score a provided batch end-to-end — cluster near-duplicates, flag duplicates, RICE or ICE score each cluster, sum requesting-customer ARR, flag backlog overlaps, and return the scored priority list.
- analyze — Health-check a request batch WITHOUT ranking it: which RICE inputs are real versus defaulted, how much of the pile is repeat tickets, zero-effort rows, unattributable ARR and objections. Deterministic, no model call.
- cluster_requests — Semantically cluster the provided requests and return the merged clusters with their members and duplicate flags — the dedupe step on its own.
- score_clusters — RICE or ICE score each merged cluster and return the scoring inputs alongside the scores.
- sum_customer_arr — Sum the requesting-customer ARR per cluster over unique customers, and return the per-cluster and total figures.
- match_backlog — Flag which clusters overlap an existing backlog item at or above the similarity threshold, returning only those overlaps.
- list_assumptions — Return the visible estimated-input assumptions the scorer made per cluster — the reach, impact, confidence and effort defaults behind each score.
- list_capabilities — List the agent's static capabilities: the available tools, the scoring models and the default similarity threshold. 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 intake form to collect the incoming requests and an optional backlog. The entry point for running the agent on your own data.
- run_form — Internal: invoked by the intake form when the user submits. Not for direct use.
