> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ada.cx/mcp/tools/get-conversations/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ada.cx/_mcp/server.

# get_conversations

Retrieves many conversations at a chosen detail level within a date range. Use this to build conversation lists for bulk analysis or to feed individual IDs into [`get_conversation`](/mcp/tools/get-conversation) for transcript review.

## Example prompts

* "Get IDs for all unresolved conversations from yesterday."
* "List conversations from last week where CSAT was 1 or 2."
* "Summarize the top themes across handoff conversations from the Voice channel in the last 7 days."

## Parameters

| Parameter      | Type   | Required | Description                                                                                                                     |
| -------------- | ------ | -------- | ------------------------------------------------------------------------------------------------------------------------------- |
| `detail_level` | string | Yes      | Detail level: `IDS_ONLY` or `SUMMARY`. See below.                                                                               |
| `start_date`   | string | Yes      | Start date in `YYYY-MM-DD` format.                                                                                              |
| `end_date`     | string | Yes      | End date in `YYYY-MM-DD` format.                                                                                                |
| `filters`      | array  | No       | Array of filter objects. Use [`get_available_filters`](/mcp/tools/get-available-filters) to construct valid filter expressions. |
| `size`         | number | No       | Maximum conversations to return. Defaults and caps vary by detail level.                                                        |
| `offset`       | number | No       | Number of conversations to skip for pagination. Must be a multiple of `size`.                                                   |

## Detail levels

* **`IDS_ONLY`** — Returns conversation ID, created date, and platform. Default 100, max 1,000. Use when you need a list of matching conversations to feed into other tools.
* **`SUMMARY`** — Returns inquiry summaries, resolution status, resolution reason, CSAT survey data (score, comment, feedback tags) when available, and a dashboard link per conversation. Default 50, max 500. Use for pattern analysis across many conversations without pulling full transcripts.