Set up CrewAI
Multi-agent orchestration framework. Add Agentcy as an MCP tool so your CrewAI agents can pull marketing data during autonomous research, analysis, and content creation tasks.
- 01
Sign up at app.goagentcy.com and create an API key
- 02
Install CrewAI with MCP support: pip install crewai 'crewai-tools[mcp]'
- 03
Configure MCPServerAdapter with the Agentcy HTTP URL
- 04
Replace YOUR_AGENTCY_API_KEY with your actual key
- 05
Pass agentcy.tools to your CrewAI agents
path: pip install 'crewai-tools[mcp]'
# Python — CrewAI with MCP
# pip install crewai 'crewai-tools[mcp]'
from crewai import Agent, Task, Crew
from crewai_tools.mcp import MCPServerAdapter
agentcy = MCPServerAdapter(
url="https://data.goagentcy.com/mcp",
headers={"Authorization": "Bearer YOUR_AGENTCY_API_KEY"},
)
agent = Agent(
role="Marketing Analyst",
goal="Analyze marketing performance data",
backstory="Expert digital marketing analyst",
tools=agentcy.tools,
)
task = Task(
description="Analyze this month's traffic and ad performance",
agent=agent,
)
crew = Crew(agents=[agent], tasks=[task])
result = crew.kickoff()CrewAI supports stdio, SSE, and Streamable HTTP transports. MCPServerAdapter provides the most control. Supports multi-server connections for combining Agentcy with other MCP tools.
“Research our competitors and create a strategy doc”
“Analyze this month’s traffic, then write a client report”
“Audit our SEO performance and suggest improvements”
“Build a content calendar based on trending keywords”