> For the complete documentation index, see [llms.txt](https://docs.darcyiq.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.darcyiq.com/grow/customer-intelligence/account-management-agent.md).

# The Account Management Agent

The **Account Management Agent** is the engine behind Customer Intelligence. Run it on a customer and it researches the account end to end — reading your CRM context, linked delivery projects, and existing intelligence, then augmenting with live web research — and writes back a refreshed account overview, health score, sentiment, signals, alerts, milestones, contacts, and revenue opportunities.

{% hint style="success" %}
**Account research on demand**: Instead of an account manager spending an afternoon piecing together the state of "Acme Corp," they click run. Minutes later the account has a current overview, a health score with rationale, three new evidence-backed signals, and two scored expansion opportunities — each citing its source.
{% endhint %}

## What the Agent Does

The agent treats each customer as an **account of your organization** and reasons about retention and growth from that perspective. In a single run it:

* Reviews the account's linked **projects** for delivery reality (status, meetings, tasks, documents)
* Reads **existing intelligence** already on file so it doesn't repeat itself
* Researches the company on the **open web** for recent, public signals (news, funding, leadership, hiring, product)
* Records **signals, alerts, and milestones** — each backed by a source
* Captures relevant **contacts** it discovers
* Writes an updated **account overview**
* Assesses **customer sentiment** (0-100 with rationale)
* Computes a **health score**
* Surfaces **revenue opportunities** with type, value, confidence, and a recommended action

## Running the Agent

The agent runs **per customer**, on demand. When triggered, it kicks off as a background task so the request returns immediately — the run continues server-side and results appear on the account as they're written.

{% hint style="info" %}
**Runs are asynchronous.** A run does real research and tool use, so it takes time. You don't need to wait on the page — refresh the account to see updated intelligence, and the account team receives a digest when the run completes (see [Notifications & Digests](/grow/customer-intelligence/notifications.md)).
{% endhint %}

## What the Agent Gathers

| Source                    | What It Provides                                                                              |
| ------------------------- | --------------------------------------------------------------------------------------------- |
| **Linked projects**       | Delivery status, meetings, tasks, documents — the engagement reality behind the relationship  |
| **Existing intelligence** | Prior signals and opportunities, so findings build on what's known                            |
| **CRM context**           | Account, contact, and pipeline data synced from your CRM                                      |
| **Web research**          | Recent public information about the company                                                   |
| **Assigned integrations** | Any integrations assigned to the customer become part of the agent's tool surface for the run |

### Project-Aware Research

If a customer has linked [Projects](/organize/projects.md), the agent can search them as first-class delivery evidence. When an account has **multiple** linked projects, the agent can search **every** linked project — selecting which project to look in as it researches — so delivery context across the whole relationship informs the account picture.

### Integration Scoping

Integrations assigned to a customer become available to the agent during the run, scoped to what the triggering user is authorized to use. This lets a run reach systems like CRM or collaboration tools where relevant, without ever exceeding the user's own permissions.

## What the Agent Produces

| Output                            | Where It Appears                                                      |
| --------------------------------- | --------------------------------------------------------------------- |
| **Account overview**              | Customer detail Overview tab                                          |
| **Health score**                  | Overview tile and the portfolio roster                                |
| **Customer sentiment**            | Overview tile (with rationale) and the roster                         |
| **Signals / Alerts / Milestones** | [Intelligence Feed](/grow/customer-intelligence/intelligence-feed.md) |
| **Contacts**                      | Connections on the account                                            |
| **Opportunities**                 | [Opportunities](/grow/customer-intelligence/opportunities.md) tab     |

How health is calculated from these outputs — and how much sentiment, alerts, milestones, delivery, and pipeline each count — is configurable. See [Health Score & Scoring Settings](/grow/customer-intelligence/settings-and-scoring.md).

## Next Steps

| Goal                            | Documentation                                                                          |
| ------------------------------- | -------------------------------------------------------------------------------------- |
| Read what the agent records     | [Intelligence Feed](/grow/customer-intelligence/intelligence-feed.md)                  |
| Work the opportunities it finds | [Opportunities](/grow/customer-intelligence/opportunities.md)                          |
| Configure how health is scored  | [Health Score & Scoring Settings](/grow/customer-intelligence/settings-and-scoring.md) |
| Learn about AI agents generally | [Agents](/core-features/agents.md)                                                     |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.darcyiq.com/grow/customer-intelligence/account-management-agent.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.