> 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.

# FAQ

**Q: What's the value of using MCP?**

A: It lets you ask natural-language questions like "How has CSAT trended this quarter?" and "What can I do to improve my CSAT?" This removes the need to dig through dashboards and helps generate insights that would be difficult to produce manually without the help of an LLM.

**Q: Can I customize which tools are available?**

A: Yes, in Claude Desktop. Once you connect the MCP server, you can disable specific tools you're not interested in using. For example, if you prefer not to share conversation transcripts, you can disable the [`get_conversation`](/mcp/tools/get-conversation) tool and still get value from the other available tools. ChatGPT does not currently support disabling individual tools.

**Q: What data flows through MCP?**

A: Depending on which tools you use, conversation transcripts, reasoning history, summaries, automated resolution status, and CSAT may be transmitted to your chosen MCP client (for example, Claude or ChatGPT). See [data privacy](/mcp/introduction/data-privacy) for the full list.

**Q: Which AI assistants are supported?**

A: Any MCP-compatible client. The [getting started](/mcp/introduction/getting-started) walk-throughs cover Claude Desktop, ChatGPT, Codex CLI, Gemini CLI, and Google ADK. Other clients like Cursor and VS Code also work—see [Other MCP clients](/mcp/introduction/getting-started/other-mcp-clients). The same tools are also available within **Microsoft Copilot**, set up as a packaged agent in Microsoft Copilot Studio rather than a direct connection—see [Other MCP clients](/mcp/introduction/getting-started/other-mcp-clients).

**Q: Does MCP support write operations?**

A: Yes, through two tools. [`edit_agent_behavior`](/mcp/tools/edit-agent-behavior) changes what the Agent knows, says, or does — knowledge articles, coaching, Playbooks, Actions, and custom instructions. Those changes are staged on a change set and go live only after explicit approval. [`edit_agent_config`](/mcp/tools/edit-agent-config) changes settings such as test cases, topics, and Custom Glossary terms, and applies immediately.

Call [`get_improvement_guide`](/mcp/tools/get-improvement-guide) first for best practices. [`send_feedback`](/mcp/tools/send-feedback) is available if a tool is blocked or missing a capability. All other tools are read-only. When using OAuth, write tools require an Owner or Admin role — see [Permissions](/mcp/introduction/authentication#permissions).

**Q: Do changes I make through MCP go live immediately?**

A: It depends on what is changed. Behavior changes are staged on a change set and stay out of live conversations until they are promoted, which requires explicit approval. Settings changes apply immediately. See [`edit_agent_behavior`](/mcp/tools/edit-agent-behavior) for the draft, test, promote, and revert flow.

**Q: Where do I see my change sets?**

A: Through [`list_agent_changesets`](/mcp/tools/list-agent-changesets) in your assistant. **Simulations** shows each test case's latest result across live and change set runs. Retrieve those results with [`get_test_runs`](/mcp/tools/get-test-runs).