> ## Documentation Index
> Fetch the complete documentation index at: https://docs.teamfollowup.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Use the public API base URL: https://api.teamfollowup.ai/api.
> Authenticate public API requests with Authorization: Bearer YOUR_API_KEY.
> Use product-owned terms: agents, campaigns, projects (GoHighLevel sub-accounts), contacts, calls, outcomes, skills, cadences, phone numbers.
> Scope requests by locationId rather than projectName.
> Read the guide linked from each API reference group before recommending an endpoint.

# Detectors

> Notice an extra condition in a call and connect it to a follow-up.

The built-in [outcomes](/agents/outcomes) cover the standard call results.
**Detectors** let you add conditions specific to your business without changing
those fixed outcomes.

A detector answers a yes-or-no question about the conversation. For example:
“Did the person mention that they need help this week?” It adds useful detail
alongside the call’s main outcome. Several detectors can be true on the same call;
each can have its own follow-up actions.

In the campaign workflow, add a detector trigger. Choose an existing detector or
create one with a name and a clear description of what it should recognize. The
software treats these detectors as yes-or-no conditions; you do not need to
choose a data type.

## Make the condition specific

“Urgent” can mean different things to different people. “The person said they
need an appointment within seven days” gives the detector something concrete to
look for. Use one condition per detector so its follow-up is easy to understand.

Creating a detector defines what to notice. Attach the desired follow-up to its
trigger to make something happen when it is detected. See
[Follow-up workflows](/campaigns/after-the-call).

## Keep existing follow-ups working

Before removing a detector, remove or change any follow-ups that use it. Renaming
or deleting a condition while a workflow still depends on it can leave that
workflow with nothing to react to.

<Note>
  The standard results and fields supplied by installed capabilities are managed
  by the software. Removing an extra detector should leave those intact.
</Note>

## In the API

* [Update agent](/api-reference/agents/update-agent) adds a detector as a `postCallFields` entry with `type: boolean`; read the list first, it is replaced whole
* [Get agent](/api-reference/agents/get-agent) reads the current list
* [Add actions to an outcome](/api-reference/campaigns/add-actions-to-an-outcome) wires the follow-up