FastGPT icon

FastGPT

AI智能客服平台

Actions37

对话 → 获取对话记录列表

AI-generated

Summary

Fetches a paginated list of individual message records within a specific FastGPT chat (conversation), optionally including user custom feedback, by POSTing to /core/chat/getPaginationRecords.

Inputs

  • 应用ID (recordsAppId) (required) — FastGPT App ID; sent in the request body as appId.
  • 历史记录ID (recordsChatId) (required) — The chatId of the conversation whose records you want; sent as chatId.
  • 偏移量 (recordsOffset) — Zero-based index of the first record to return; default 0. Sent as offset.
  • 每页数量 (recordsPageSize) — Number of records per page; default 10. Sent as pageSize.
  • 读取自定义反馈 (recordsLoadCustomFeedbacks) — Whether to load user custom feedback info into the records; default true. Sent as loadCustomFeedbacks.
  • Request Options — Optional node settings: batching (items per batch, interval), ignore SSL issues, proxy, and timeout (default 10000 ms).

Output shape

The FastGPT API JSON response for the page of chat records (the raw response body, not unwrapped), one output per input item.

The exact record fields come from the FastGPT server and aren't defined in the node. HTTP 4xx/5xx statuses or a body code other than 200 make the node throw an API error that carries the server's message. To page through a long chat, increase offset by pageSize on each call.

Examples

Example 1: Read the first 10 messages of a known conversation

应用ID = your App ID, 历史记录ID = the chatId, offset 0, page size 10, feedbacks on (defaults).

Example 2: Fetch the next page of a long conversation without feedback data

Same App ID and chatId, offset 10, page size 10, 读取自定义反馈 off.

Example 3: Pull transcripts after listing histories

Run 获取历史记录列表 first, then map each item's chatId via expression into 历史记录ID here.

Links

Discussion