Tool Types
Bananaflow provides two categories of tools:Built-in Tools
Pre-configured tools that handle common telephony operations out of the box:- Call Transfer – Transfer the active call to a phone number or SIP endpoint
- Transfer to Agent – Hand the live call to another agent, carrying conversation history as a labeled transcript
- End Call – Terminate the call when the conversation is complete
Custom Tools
Tools you define to integrate with any external system:- HTTP API – Call any REST API endpoint during a conversation (e.g., CRM updates, data lookups, triggering automations)
- MCP Tool – Connect an external MCP server and expose its remote tools to the LLM during a conversation
- Calculator – Do arithmetic during the call. It supports
+,-,*,/,**,%and parentheses. - Integration – Run an action in a connected app, like sending a Gmail email or creating a Google Calendar event. See Integration tools.
- Reasoner – Hand one hard question, like a bill total or a policy rule, to a slower model that works it out while the agent keeps talking. You do not create it in the Create New Tool dialog. Ask your AI assistant to add it through the MCP server.
Integration tools
An integration tool runs one action in an app you have connected, for example “create a contact” in HubSpot or “send an email” in Gmail.- On the Tools page, click Create Tool and choose Integration.
- Search for the app, like Gmail or Slack, then choose an action.
- If the app is not connected yet, a sign-in window opens. Finish the sign-in there.
- Fill in the arguments and save.
Manage your tools
The Tools page has three tabs:- Tools – your active tools.
- Archived – tools you archived. You can bring one back.
- Connections – the apps behind your integration tools.
How Tools Work
- You define a tool with a name, description, and parameters
- You attach the tool to one or more workflow nodes
- During a call, the LLM reads your node prompt, the tool description, and the caller’s intent to decide whether to invoke the tool
- The tool executes and returns a result that the agent can use to continue the conversation
Best Practices
- Attach only relevant tools to each node – fewer tools means more reliable invocations
- Write clear tool descriptions – the LLM uses these to decide when to call the tool
- Guide the LLM in your node prompt – explicitly describe when a tool should be used
- Test tool behavior – verify your agent invokes tools at the right moments using web or phone calls, or save the check as an eval