---
title: "Get Suggested Response Prompt"
method: GET
path: "/calls/{call_id}/suggested-response-prompt"
tags: ["calls"]
---

# Get Suggested Response Prompt

`GET /calls/{call_id}/suggested-response-prompt`

## Request

- Base URL: `https://dev.whispa.net` (the document declares no server; this is the origin it was published from)
- URL: `https://dev.whispa.net/calls/{call_id}/suggested-response-prompt`
- Auth: one of:
  - OAuth 2
    - Password
    - Token URL: `auth/bearer/login`
  - API key in cookie `access_token`

Rendered prompt + output for the suggested response at a transcript turn.

Reconstructed from the Langfuse generation trace for prompt tuning (compare
what we suggested, and the context we fed in, against the agent's real
reply). Owner-or-admin only, same as replay.

## Path parameters

- `call_id` string, uuid, required

## Query parameters

- `sequence` integer, required

## Response `200`

Successful Response

- SuggestedResponsePromptResponse: Rendered prompt + output for the suggested response at one turn. Reconstructed from the Langfuse generation trace so the suggested-responses prompt can be tuned against what the agent actually said. Output-only data lives in our DB; the prompt *input* (context) only exists in Langfuse.
  - `found` boolean, required: Whether a matching Langfuse generation was found
  - `sequence` integer, required: Transcript sequence of the turn
  - `system_prompt` string, nullable
  - `user_prompt` string, nullable: Rendered user prompt incl. injected context
  - `output` string, nullable: The suggestion the model produced
  - `trace_url` string, nullable: Deep link to the Langfuse trace
  - `detail` string, nullable: Reason no prompt was returned, when found is false

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/whispa/apis/whispa-api.md) - [All operations](https://skmtc.dev/whispa/apis/whispa-api/llms.txt) - [OpenAPI document](https://skmtc.dev/whispa/apis/whispa-api/revisions/12c2e4aca43c?raw)