Tool-Call Transparency UI

Surface an agent's background tool activity in the chat UI itself so users can inspect the rationale behind an unexpected response.

The turn-taking chat UI has stayed essentially unchanged for decades; the affordances that need to be added on top are specific to function calling agents:

  1. Processing spinner — indicates the agent is working and a response is forthcoming.
  2. Tool-use indicator — a small marker inside the agent's message (e.g. a pill button) signaling background tool activity is occurring or occurred.
  3. Tool-call detail view — a "Tool calls" control revealing full details (tool name, arguments as a webform, results) so the user can inspect the rationale behind a response, especially when it's unexpected.

Detail-view visibility matters beyond UX polish: it is what makes editable tool-call correction possible, and it gives users the same parameter-level view that human approval gates recommend logging for reviewers.

Sources

  • Prompt Engineering for LLMs (John Berryman), ch. 8