Explore Custom Tools

Explore Custom Tools

What “Custom Tools” mean in FlowbotAI

Custom tools give your FlowbotAI agent the ability to communicate with systems outside the call—anything you can do in a function (or API endpoint) can be exposed to the agent as a tool. Tools are the foundation for practical automation like sending notifications, retrieving customer details, updating records, and more.

What custom tools unlock

  • External actions: trigger workflows like sending emails, updating records, or creating tickets.
  • External lookups: fetch data from CRMs, billing systems, or internal APIs at runtime.
  • Real-world integration: connect your agent to any HTTP service or client-implemented function your app provides.


How FlowbotAI passes inputs to custom tools

Tool parameters define what your agent sends to your backend implementation when it invokes a tool. A clean parameter design makes tool calls more reliable, reduces “guessing,” and keeps conversations moving at a natural pace. When you create a custom tool, every parameter belongs to one of three buckets:

  1. Dynamic (chosen by the model),
  2. Static (hard-coded at definition time)
  3. Automatic (filled in by the platform at runtime)

Choosing the right parameter type (rule of thumb):
  • If the caller might say it (or you might ask for it), make it Dynamic.
  • If it should never change for this tool, make it Static.
  • If FlowbotAI can supply it reliably at runtime, make it Automatic.

Dynamic parameters (agent-generated inputs at runtime)

Dynamic parameters have their values generated by the model at the moment the tool is called. Use them for things like customer identifiers, order numbers, and any other caller-driven inputs.


Static parameters (fixed values the model never sees)

Static parameters are known at tool creation time and are always sent with every invocation. Static parameters are not exposed to (and cannot be set by) the model.


Automatic parameters (FlowbotAI supplies known runtime context)

Automatic parameters are used when you want a predictable value that is not generated by the model, but the value isn’t known when the tool is created. FlowbotAI populates these values automatically at runtime.

Common automatic values you can request

knownValue
What you get
KNOWN_PARAM_CALL_ID
Sends the current FlowbotAI call ID to your tool.
KNOWN_PARAM_CONVERSATION_HISTORYIncludes the full conversation history leading up to the tool call (commonly sent in the request body).
KNOWN_PARAM_CALL_STATEIncludes arbitrary state previously set by tools.



Supported Authentication Methods

FlowbotAI supports three methods for attaching API keys to tool HTTP requests:
  1. Query parameter: The key is added to the URL query string. Use this if the downstream service expects an API key in the query string.
  2. Custom header: The key is sent in a header you name. se this if the downstream service expects an API key in a custom HTTP header.
  3. Authorization header: The key is sent using an auth scheme (for example, Bearer). Use this if the downstream service expects credentials via the Authorization header.

Support for multiple authentication options

A single tool can declare multiple authentication options. This is useful if your server accepts more than one method (for example, either query-string auth or header auth).

Each option can also include multiple requirements, such as requiring both a user identifier and an auth token.

If the tokens you provide satisfy multiple options, FlowbotAI uses the first non-empty option whose requirements are all satisfied.

If an unauthenticated option is present, it will only be used if no other option can be satisfied.


Info
Quick validation checklist
  1. Confirm your tool definition declares the expected auth method (query parameter name, header name, or Authorization scheme).
  2. Make sure the authTokens keys you provide match the tool’s declared requirements (example key name: myServiceApiKey).
  3. Verify your backend is reading the key from the same location you configured (query string vs header vs Authorization).
  4. If multiple auth options are defined, verify at least one complete option can be satisfied by the tokens you provided; otherwise an unauthenticated fallback (if present) may be chosen.





    • Related Articles

    • Explore Built-in Tools

      What “built-in tools” mean in FlowbotAI FlowbotAI ships with a small set of built-in tools that cover common voice-application needs out of the box. They behave the same way as tools you create yourself: your agent can invoke them mid-conversation, ...

    • FlowbotAI Tools Overview

      What tools unlock in FlowbotAI Tools are the “action layer” for FlowbotAI voice agents. They let an agent do more than talk — it can look things up, update systems, and trigger workflows while keeping the conversation natural and on-track. In plain ...

    • Zapier Worfkflow Integration

      Flowbot Agent Connector – Zapier Integration Guide The Flowbot Agent Connector allows you to integrate your Flowbot Voice Agents with over 8,000+ apps in the Zapier ecosystem. Automate everyday tasks like creating tickets, updating CRMs, sending ...

    • Knowledge (RAG) Overview

      Give your agent access to your KB, product docs, playbooks, and internal procedures—without stuffing everything into the prompt. Why RAG matters (especially in production) Large language models know a lot about the world, but they won’t reliably know ...

    • Agent Reaction After Tool Calls

      FlowbotAI Agent Reactions After Tool Calls When an agent calls a tool, you can control what happens next: should the agent speak immediately, stay silent and listen, or speak only in specific situations. This helps you avoid awkward “double talk” ...