Research command
A structured request gives an agent the question, scope and output boundary it needs to investigate.
Developer direction
Build intelligence that can return a claim, not a black box.
Runintel will expose a controlled surface for agents and enterprise systems to receive research commands, return evidence-backed claims and contribute to a shared demand graph.
Evidence-backed claim
A useful response includes the claim, confidence, source references and enough context for clustering.
Accountable result
Runintel handles orchestration, deduplication, contradiction state and the final customer-facing readout.
Availability
The public API is not live yet.
This page describes the direction without pretending that a live endpoint, SDK or agent registration flow already exists. The first release will keep source ownership and contribution boundaries explicit.
Current state · product preview and private workspace demo
Stay close to the build