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

# MCP server: AI Agent and human agent CSAT in get_ada_metric

[`get_ada_metric`](/mcp/tools/get-ada-metric) has two new metric types, `csat_rate_ai` and `csat_rate_human`. They return the **AI Agent Score** and the **Human Agent Score** from the **Customer satisfaction score** report. Each survey counts for the AI Agent or for the human agent who joined the conversation. The two metric types return `null` when no survey matches.

## Change to the HANDOFF filter

`csat_rate` with the `HANDOFF` filter now returns the overall CSAT of the conversations that the filter matches. Before this change, `csat_rate` with `HANDOFF` returned the AI Agent or human agent score and did not filter conversations. The response has a `note` field when `csat_rate` comes with `HANDOFF`.

An assistant that used `csat_rate` with `HANDOFF` to get the AI Agent or human agent score must change to `csat_rate_ai` or `csat_rate_human`. Combine a metric type with `HANDOFF` to match the report with a Handoff filter. For example, `csat_rate_ai` with `HANDOFF` `ISNOT` `true` matches the **AI Agent Score** with **Handoff is not true**.

`get_ada_metric` reads dates in UTC. The report reads dates in the time zone of the viewer. If your time zone is not UTC, the numbers can differ by the surveys near the start and end dates.