> For a complete page index, fetch https://docs.synthflow.ai/llms.txt. For full documentation content, fetch https://docs.synthflow.ai/llms-full.txt.

# Anthropic

> Connect Claude or Claude Code to Synthflow with the Synthflow connector to build agents, review calls, and manage knowledge bases in plain language.

![Synthflow AI Anthropic connector](https://storage.googleapis.com/granular-changelog/doc-images/anthropic_cover.svg)

Anthropic's **Claude** and **Claude Code** both let you run your Synthflow workspace from a conversation. Connect the Synthflow connector once, then create and manage agents, place and review calls, build knowledge bases, and run actions by describing what you want in plain language, without leaving the tool you already use. The connector works the same way in both.

This is a good fit when your work already lives in Claude.

## What you can do

![Synthflow tool permissions in the Claude connector](https://storage.googleapis.com/granular-changelog/doc-images/claude_tools.png)

Ask Claude to:

* **Build and update agents**: Create agents, draft and refine prompts, publish drafts, and compare versions before you go live.
* **Manage calls**: Place test calls, review recent calls, and surface summaries or negative-sentiment moments when you ask.
* **Manage knowledge**: Create knowledge bases, add items from files or URLs, and attach them to the agents that should use them.
* **Run actions and simulations**: Scaffold and test [actions](/actions-overview), then run [simulations](/simulations) to validate behavior before launch.

Synthflow exposes these capabilities as individual tools, grouped into **read-only tools** that look things up (such as listing agents or reviewing calls) and **write and delete tools** that change your workspace (such as creating an agent or removing a knowledge base item). Claude lets you set an approval policy for each group, so you control what runs automatically and what waits for your confirmation. Keep write and delete tools behind approval unless you are sure. For the full list of tools, see the [capabilities table](/aurora#capabilities).

## Setup instructions

The Synthflow connector works the same way in both Claude and Claude Code. You sign in through Synthflow, so there is no API key to copy or manage by hand. Know which [region](/customer-region) your workspace belongs to, then connect the Synthflow connector to the matching MCP server URL:

| Region         | MCP URL                           |
| -------------- | --------------------------------- |
| Global         | `https://mcp.synthflow.ai/mcp`    |
| United States  | `https://mcp.us.synthflow.ai/mcp` |
| European Union | `https://mcp.eu.synthflow.ai/mcp` |

The server URL and your workspace must be in the same [region](/customer-region). A mismatch, such as the EU URL with a US workspace, stops the connection from working.

You are redirected to Synthflow to log in, where you select the workspace you want to manage.

![Selecting a Synthflow workspace during connection](https://storage.googleapis.com/granular-changelog/doc-images/synthflow_workspace_selector.png)

After you select the workspace, click **Allow access** on the next screen and the connection is established. You can then start prompting.

## Example prompts

| Example prompt                                                                  |
| :------------------------------------------------------------------------------ |
| List my agents and summarize what each one does.                                |
| Create a new inbound sales qualification agent and publish it.                  |
| Show me the last 20 calls and flag any with negative sentiment.                 |
| Create a knowledge base from these sources, then attach it to my support agent. |
| Run a simulation where a prospect asks about pricing and implementation time.   |

## Security considerations

Connecting an AI tool to operational tools introduces new risks. The primary risk class unique to these workflows is **prompt injection**: untrusted content (like call transcripts, user messages, or knowledge base documents) can include instructions that try to trick the model into taking unintended actions. Keep **confirm before running tools** enabled and review each action carefully.

## FAQ

No. Both use the same Synthflow connector: you sign in through Synthflow and pick a workspace, with no API key to manage. See [Setup instructions](#setup-instructions).

[Aurora](/aurora) is the in-dashboard agent: one chat thread in the Synthflow UI with Memory, Learnings, and rich UI cards. The Synthflow connector runs inside Claude or Claude Code and manages the workspace you connect. All draw on the same [capabilities](/aurora#capabilities).

[MCP Connectors](/mcp-actions) let a **voice agent** call tools on an **external** server during a live call. Claude and Claude Code are the opposite direction: they call tools to manage **your Synthflow workspace**.

Claude and Claude Code run in your own Anthropic account, so model and subscription charges are billed by Anthropic according to your plan. Synthflow bills only for the workspace usage your actions generate, such as calls, under your existing Synthflow plan.