Skip to content

Commit 51eb5e9

Browse files
docs: add CrewAI Enterprise docs (#2691)
* Add enterprise deployment documentation to CLI docs * Update CrewAI Enterprise documentation with comprehensive guides for Traces, Tool Repository, Webhook Streaming, and FAQ structure * Add Enterprise documentation images * Update Enterprise introduction with visual CardGroups and Steps components
1 parent b2969e9 commit 51eb5e9

37 files changed

+1637
-0
lines changed

docs/docs.json

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -180,6 +180,42 @@
180180
}
181181
]
182182
},
183+
{
184+
"tab": "Enterprise",
185+
"groups": [
186+
{
187+
"group": "Getting Started",
188+
"pages": [
189+
"enterprise/introduction"
190+
]
191+
},
192+
{
193+
"group": "How-To Guides",
194+
"pages": [
195+
"enterprise/guides/build-crew",
196+
"enterprise/guides/deploy-crew",
197+
"enterprise/guides/kickoff-crew",
198+
"enterprise/guides/update-crew",
199+
"enterprise/guides/use-crew-api",
200+
"enterprise/guides/enable-crew-studio"
201+
]
202+
},
203+
{
204+
"group": "Features",
205+
"pages": [
206+
"enterprise/features/tool-repository",
207+
"enterprise/features/webhook-streaming",
208+
"enterprise/features/traces"
209+
]
210+
},
211+
{
212+
"group": "Resources",
213+
"pages": [
214+
"enterprise/resources/frequently-asked-questions"
215+
]
216+
}
217+
]
218+
},
183219
{
184220
"tab": "Examples",
185221
"groups": [
Lines changed: 106 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,106 @@
1+
---
2+
title: Tool Repository
3+
description: "Using the Tool Repository to manage your tools"
4+
icon: "toolbox"
5+
---
6+
7+
## Overview
8+
9+
The Tool Repository is a package manager for CrewAI tools. It allows users to publish, install, and manage tools that integrate with CrewAI crews and flows.
10+
11+
Tools can be:
12+
13+
- **Private**: accessible only within your organization (default)
14+
- **Public**: accessible to all CrewAI users if published with the `--public` flag
15+
16+
The repository is not a version control system. Use Git to track code changes and enable collaboration.
17+
18+
## Prerequisites
19+
20+
Before using the Tool Repository, ensure you have:
21+
22+
- A [CrewAI Enterprise](https://app.crewai.com) account
23+
- [CrewAI CLI](https://docs.crewai.com/concepts/cli#cli) installed
24+
- [Git](https://git-scm.com) installed and configured
25+
- Access permissions to publish or install tools in your CrewAI Enterprise organization
26+
27+
## Installing Tools
28+
29+
To install a tool:
30+
31+
```bash
32+
crewai tool install <tool-name>
33+
```
34+
35+
This installs the tool and adds it to `pyproject.toml`.
36+
37+
## Creating and Publishing Tools
38+
39+
To create a new tool project:
40+
41+
```bash
42+
crewai tool create <tool-name>
43+
```
44+
45+
This generates a scaffolded tool project locally.
46+
47+
After making changes, initialize a Git repository and commit the code:
48+
49+
```bash
50+
git init
51+
git add .
52+
git commit -m "Initial version"
53+
```
54+
55+
To publish the tool:
56+
57+
```bash
58+
crewai tool publish
59+
```
60+
61+
By default, tools are published as private. To make a tool public:
62+
63+
```bash
64+
crewai tool publish --public
65+
```
66+
67+
For more details on how to build tools, see [Creating your own tools](https://docs.crewai.com/concepts/tools#creating-your-own-tools).
68+
69+
## Updating Tools
70+
71+
To update a published tool:
72+
73+
1. Modify the tool locally
74+
2. Update the version in `pyproject.toml` (e.g., from `0.1.0` to `0.1.1`)
75+
3. Commit the changes and publish
76+
77+
```bash
78+
git commit -m "Update version to 0.1.1"
79+
crewai tool publish
80+
```
81+
82+
## Deleting Tools
83+
84+
To delete a tool:
85+
86+
1. Go to [CrewAI Enterprise](https://app.crewai.com)
87+
2. Navigate to **Tools**
88+
3. Select the tool
89+
4. Click **Delete**
90+
91+
<Warning>
92+
Deletion is permanent. Deleted tools cannot be restored or re-installed.
93+
</Warning>
94+
95+
## Security Checks
96+
97+
Every published version undergoes automated security checks, and are only available to install after they pass.
98+
99+
You can check the security check status of a tool at:
100+
101+
`CrewAI Enterprise > Tools > Your Tool > Versions`
102+
103+
104+
<Card title="Need Help?" icon="headset" href="mailto:support@crewai.com">
105+
Contact our support team for assistance with API integration or troubleshooting.
106+
</Card>

docs/enterprise/features/traces.mdx

Lines changed: 146 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,146 @@
1+
---
2+
title: Traces
3+
description: "Using Traces to monitor your Crews"
4+
icon: "timeline"
5+
---
6+
7+
## Overview
8+
9+
Traces provide comprehensive visibility into your crew executions, helping you monitor performance, debug issues, and optimize your AI agent workflows.
10+
11+
## What are Traces?
12+
13+
Traces in CrewAI Enterprise are detailed execution records that capture every aspect of your crew's operation, from initial inputs to final outputs. They record:
14+
15+
- Agent thoughts and reasoning
16+
- Task execution details
17+
- Tool usage and outputs
18+
- Token consumption metrics
19+
- Execution times
20+
- Cost estimates
21+
22+
<Frame>
23+
![Traces Overview](/images/enterprise/traces-overview.png)
24+
</Frame>
25+
26+
## Accessing Traces
27+
28+
<Steps>
29+
<Step title="Navigate to the Traces Tab">
30+
Once in your CrewAI Enterprise dashboard, click on the **Traces** to view all execution records.
31+
</Step>
32+
33+
<Step title="Select an Execution">
34+
You'll see a list of all crew executions, sorted by date. Click on any execution to view its detailed trace.
35+
</Step>
36+
</Steps>
37+
38+
## Understanding the Trace Interface
39+
40+
The trace interface is divided into several sections, each providing different insights into your crew's execution:
41+
42+
### 1. Execution Summary
43+
44+
The top section displays high-level metrics about the execution:
45+
46+
- **Total Tokens**: Number of tokens consumed across all tasks
47+
- **Prompt Tokens**: Tokens used in prompts to the LLM
48+
- **Completion Tokens**: Tokens generated in LLM responses
49+
- **Requests**: Number of API calls made
50+
- **Execution Time**: Total duration of the crew run
51+
- **Estimated Cost**: Approximate cost based on token usage
52+
53+
<Frame>
54+
![Execution Summary](/images/enterprise/trace-summary.png)
55+
</Frame>
56+
57+
### 2. Tasks & Agents
58+
59+
This section shows all tasks and agents that were part of the crew execution:
60+
61+
- Task name and agent assignment
62+
- Agents and LLMs used for each task
63+
- Status (completed/failed)
64+
- Individual execution time of the task
65+
66+
<Frame>
67+
![Task List](/images/enterprise/trace-tasks.png)
68+
</Frame>
69+
70+
### 3. Final Output
71+
72+
Displays the final result produced by the crew after all tasks are completed.
73+
74+
<Frame>
75+
![Final Output](/images/enterprise/final-output.png)
76+
</Frame>
77+
78+
### 4. Execution Timeline
79+
80+
A visual representation of when each task started and ended, helping you identify bottlenecks or parallel execution patterns.
81+
82+
<Frame>
83+
![Execution Timeline](/images/enterprise/trace-timeline.png)
84+
</Frame>
85+
86+
### 5. Detailed Task View
87+
88+
When you click on a specific task in the timeline or task list, you'll see:
89+
90+
<Frame>
91+
![Detailed Task View](/images/enterprise/trace-detailed-task.png)
92+
</Frame>
93+
94+
- **Task Key**: Unique identifier for the task
95+
- **Task ID**: Technical identifier in the system
96+
- **Status**: Current state (completed/running/failed)
97+
- **Agent**: Which agent performed the task
98+
- **LLM**: Language model used for this task
99+
- **Start/End Time**: When the task began and completed
100+
- **Execution Time**: Duration of this specific task
101+
- **Task Description**: What the agent was instructed to do
102+
- **Expected Output**: What output format was requested
103+
- **Input**: Any input provided to this task from previous tasks
104+
- **Output**: The actual result produced by the agent
105+
106+
107+
## Using Traces for Debugging
108+
109+
Traces are invaluable for troubleshooting issues with your crews:
110+
111+
<Steps>
112+
<Step title="Identify Failure Points">
113+
When a crew execution doesn't produce the expected results, examine the trace to find where things went wrong. Look for:
114+
115+
- Failed tasks
116+
- Unexpected agent decisions
117+
- Tool usage errors
118+
- Misinterpreted instructions
119+
120+
<Frame>
121+
![Failure Points](/images/enterprise/failure.png)
122+
</Frame>
123+
</Step>
124+
125+
<Step title="Optimize Performance">
126+
Use execution metrics to identify performance bottlenecks:
127+
128+
- Tasks that took longer than expected
129+
- Excessive token usage
130+
- Redundant tool operations
131+
- Unnecessary API calls
132+
</Step>
133+
134+
<Step title="Improve Cost Efficiency">
135+
Analyze token usage and cost estimates to optimize your crew's efficiency:
136+
137+
- Consider using smaller models for simpler tasks
138+
- Refine prompts to be more concise
139+
- Cache frequently accessed information
140+
- Structure tasks to minimize redundant operations
141+
</Step>
142+
</Steps>
143+
144+
<Card title="Need Help?" icon="headset" href="mailto:support@crewai.com">
145+
Contact our support team for assistance with trace analysis or any other CrewAI Enterprise features.
146+
</Card>
Lines changed: 82 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,82 @@
1+
---
2+
title: Webhook Streaming
3+
description: "Using Webhook Streaming to stream events to your webhook"
4+
icon: "webhook"
5+
---
6+
7+
## Overview
8+
9+
Enterprise Event Streaming lets you receive real-time webhook updates about your crews and flows deployed to
10+
CrewAI Enterprise, such as model calls, tool usage, and flow steps.
11+
12+
## Usage
13+
14+
When using the Kickoff API, include a `webhooks` object to your request, for example:
15+
16+
```json
17+
{
18+
"inputs": {"foo": "bar"},
19+
"webhooks": {
20+
"events": ["crew_kickoff_started", "llm_call_started"],
21+
"url": "https://your.endpoint/webhook",
22+
"realtime": false,
23+
"authentication": {
24+
"strategy": "bearer",
25+
"token": "my-secret-token"
26+
}
27+
}
28+
}
29+
```
30+
31+
If `realtime` is set to `true`, each event is delivered individually and immediately, at the cost of crew/flow performance.
32+
33+
## Webhook Format
34+
35+
Each webhook sends a list of events:
36+
37+
```json
38+
{
39+
"events": [
40+
{
41+
"id": "event-id",
42+
"execution_id": "crew-run-id",
43+
"timestamp": "2025-02-16T10:58:44.965Z",
44+
"type": "llm_call_started",
45+
"data": {
46+
"model": "gpt-4",
47+
"messages": [
48+
{"role": "system", "content": "You are an assistant."},
49+
{"role": "user", "content": "Summarize this article."}
50+
]
51+
}
52+
}
53+
]
54+
}
55+
```
56+
57+
The `data` object structure varies by event type. Refer to the [event list](https://github.com/crewAIInc/crewAI/tree/main/src/crewai/utilities/events) on GitHub.
58+
59+
As requests are sent over HTTP, the order of events can't be guaranteed. If you need ordering, use the `timestamp` field.
60+
61+
## Supported Events
62+
63+
CrewAI supports both system events and custom events in Enterprise Event Streaming. These events are sent to your configured webhook endpoint during crew and flow execution.
64+
65+
- `crew_kickoff_started`
66+
- `crew_step_started`
67+
- `crew_step_completed`
68+
- `crew_execution_completed`
69+
- `llm_call_started`
70+
- `llm_call_completed`
71+
- `tool_usage_started`
72+
- `tool_usage_completed`
73+
- `crew_test_failed`
74+
- *...and others*
75+
76+
Event names match the internal event bus. See [GitHub source](https://github.com/crewAIInc/crewAI/tree/main/src/crewai/utilities/events) for the full list.
77+
78+
You can emit your own custom events, and they will be delivered through the webhook stream alongside system events.
79+
80+
<Card title="Need Help?" icon="headset" href="mailto:support@crewai.com">
81+
Contact our support team for assistance with webhook integration or troubleshooting.
82+
</Card>

0 commit comments

Comments
 (0)