Run steps
The history of each tool call: arguments, result, and tokens.
Updated on Aug 09, 2026
Each run keeps the full trail: which tool was called, with which arguments, what came back, and how many tokens each step cost. That is what lets you debug an agent that took a strange path — and audit what it did in production.
GET
https://api.hinow.ai/v1/agents/{id}/runs/{runId}/stepsBearerStep history: each tool call, with arguments, result, and tokens.
Parâmetros
idstring· pathobrigatóriorunIdstring· pathobrigatório
Respostas
200List of steps, in order
No native tools?
Models without function calling support are still usable: agent-kit brings an XML tag protocol that runs in the same loop, without the tools field.
Was this page helpful?

