Ask CrewAI Docs
Use the live CrewAI documentation to answer questions with up-to-date, authoritative information.
When to Use This Skill
Use this skill when:
- The user asks about a CrewAI feature, parameter, or behavior not covered in detail by the other skills
- You need to verify current API syntax, method signatures, or configuration options
- The user hits an error and needs troubleshooting guidance from official docs
- The question is about a newer or less common CrewAI feature (e.g., telemetry, testing, CLI commands, deployment, enterprise features)
- You're unsure whether your knowledge is current — the docs reflect the latest published state
Do NOT use this skill when the question is clearly answered by one of the other skills (getting-started, design-agent, design-task, optimize-flow). Those skills contain curated, opinionated guidance. This skill is for filling gaps and verifying details.
How to Query the Docs
Try the approaches below in order. Use the first one that's available.
Option 1: CrewAI Docs MCP Server (Preferred)
If the crewai-docs MCP server is configured, use its tools directly to search and read documentation. This is the best experience — structured search with full page content.
Option 2: WebFetch Fallback
If the MCP server is not configured, fall back to fetching docs via the web:
Find the right page — fetch the docs index to locate the relevant page:
WebFetch: https://docs.crewai.com/llms.txt
This returns a sitemap of all doc pages with descriptions. Identify the URL most relevant to the user's question.
Fetch the page — retrieve the specific doc page content:
WebFetch: https://docs.crewai.com/<path-from-index>
Synthesize the answer — combine what you find with context from the other skills to give a clear, actionable response.
Cite the source — include the docs URL so the user can read further.
After using the fallback, suggest the user configure the MCP server for a better experience:
Tip: For faster docs lookups, add the CrewAI docs MCP server to your coding agent:
https://docs.crewai.com/mcp
Setting Up the MCP Server (Recommended)
For the best experience, configure the CrewAI documentation MCP server in your coding agent.
Server URL:
https://docs.crewai.com/mcp
Codex
Add to .Codex/settings.json (project-level) or ~/.Codex/settings.json (global):
{
"mcpServers": {
"crewai-docs": {
"type": "url",
"url": "https://docs.crewai.com/mcp"
}
}
}
Cursor / Windsurf / Other Agents
Add https://docs.crewai.com/mcp as a remote MCP server following your tool's MCP configuration docs.
Workflow
- Understand the user's question — what specific CrewAI concept, API, or behavior are they asking about?
- Query the docs — use the MCP tools if available, otherwise WebFetch the relevant page
- Synthesize the answer — combine what you find from the docs with context from the other skills to give a clear, actionable response
- Cite the source — mention which docs page the information came from so the user can read further
Examples of Good Use Cases
| User Question |
Why This Skill |
"What parameters does Crew() accept?" |
Specific API reference — docs are authoritative |
| "How do I set up telemetry in CrewAI?" |
Niche feature not covered in other skills |
"What's the difference between Process.sequential and Process.hierarchical?" |
Detailed comparison best sourced from docs |
"I'm getting ValidationError when using output_pydantic" |
Troubleshooting — docs may have known issues or caveats |
| "How do I deploy a CrewAI flow to production?" |
Deployment guidance lives in docs, not in design skills |
"What CLI commands does crewai support?" |
CLI reference is a docs concern |
| "How do I configure memory for a crew?" |
Detailed config options beyond what design-agent covers |
Related Skills
- getting-started — project scaffolding, choosing abstractions, Flow architecture
- design-agent — agent Role-Goal-Backstory, parameter tuning, tools, memory & knowledge
- design-task — task descriptions, expected_output, guardrails, structured output, dependencies
- optimize-flow — Flow latency optimization, parallelization, model tiering
1---2name: ask-docs3description: Query the official CrewAI documentation via its live MCP server. Use when the user has a CrewAI question that isn't fully covered by the getting-started, design-agent, design-task, or optimize-flow skills — e.g., specific API details, configuration options, advanced features, troubleshooting errors, or anything where the latest docs are the best source of truth.4---5
6# Ask CrewAI Docs
7
8Use the live CrewAI documentation to answer questions with up-to-date, authoritative information.
9
10---
11
12## When to Use This Skill
13
14Use this skill when:
15
16- The user asks about a CrewAI feature, parameter, or behavior not covered in detail by the other skills
17- You need to verify current API syntax, method signatures, or configuration options
18- The user hits an error and needs troubleshooting guidance from official docs
19- The question is about a newer or less common CrewAI feature (e.g., telemetry, testing, CLI commands, deployment, enterprise features)
20- You're unsure whether your knowledge is current — the docs reflect the latest published state
21
22**Do NOT use this skill** when the question is clearly answered by one of the other skills (getting-started, design-agent, design-task, optimize-flow). Those skills contain curated, opinionated guidance. This skill is for filling gaps and verifying details.
23
24---
25
26## How to Query the Docs
27
28Try the approaches below in order. Use the first one that's available.
29
30### Option 1: CrewAI Docs MCP Server (Preferred)
31
32If the `crewai-docs` MCP server is configured, use its tools directly to search and read documentation. This is the best experience — structured search with full page content.
33
34### Option 2: WebFetch Fallback
35
36If the MCP server is not configured, fall back to fetching docs via the web:
37
381. **Find the right page** — fetch the docs index to locate the relevant page:
39 ```
40 WebFetch: https://docs.crewai.com/llms.txt
41 ```
42 This returns a sitemap of all doc pages with descriptions. Identify the URL most relevant to the user's question.
43
442. **Fetch the page** — retrieve the specific doc page content:
45 ```
46 WebFetch: https://docs.crewai.com/<path-from-index>
47 ```
48
493. **Synthesize the answer** — combine what you find with context from the other skills to give a clear, actionable response.
50
514. **Cite the source** — include the docs URL so the user can read further.
52
53After using the fallback, suggest the user configure the MCP server for a better experience:
54
55> **Tip:** For faster docs lookups, add the CrewAI docs MCP server to your coding agent:
56> `https://docs.crewai.com/mcp`
57
58---
59
60## Setting Up the MCP Server (Recommended)
61
62For the best experience, configure the CrewAI documentation MCP server in your coding agent.
63
64**Server URL:**
65```
66https://docs.crewai.com/mcp
67```
68
69### Codex
70
71Add to `.Codex/settings.json` (project-level) or `~/.Codex/settings.json` (global):
72
73```json
74{
75 "mcpServers": {
76 "crewai-docs": {
77 "type": "url",
78 "url": "https://docs.crewai.com/mcp"
79 }
80 }
81}
82```
83
84### Cursor / Windsurf / Other Agents
85
86Add `https://docs.crewai.com/mcp` as a remote MCP server following your tool's MCP configuration docs.
87
88---
89
90## Workflow
91
921. **Understand the user's question** — what specific CrewAI concept, API, or behavior are they asking about?
932. **Query the docs** — use the MCP tools if available, otherwise WebFetch the relevant page
943. **Synthesize the answer** — combine what you find from the docs with context from the other skills to give a clear, actionable response
954. **Cite the source** — mention which docs page the information came from so the user can read further
96
97---
98
99## Examples of Good Use Cases
100
101| User Question | Why This Skill |
102|---|---|
103| "What parameters does `Crew()` accept?" | Specific API reference — docs are authoritative |
104| "How do I set up telemetry in CrewAI?" | Niche feature not covered in other skills |
105| "What's the difference between `Process.sequential` and `Process.hierarchical`?" | Detailed comparison best sourced from docs |
106| "I'm getting `ValidationError` when using `output_pydantic`" | Troubleshooting — docs may have known issues or caveats |
107| "How do I deploy a CrewAI flow to production?" | Deployment guidance lives in docs, not in design skills |
108| "What CLI commands does `crewai` support?" | CLI reference is a docs concern |
109| "How do I configure memory for a crew?" | Detailed config options beyond what design-agent covers |
110
111---
112
113## Related Skills
114
115- **getting-started** — project scaffolding, choosing abstractions, Flow architecture
116- **design-agent** — agent Role-Goal-Backstory, parameter tuning, tools, memory & knowledge
117- **design-task** — task descriptions, expected_output, guardrails, structured output, dependencies
118- **optimize-flow** — Flow latency optimization, parallelization, model tiering
crewaiinc