Orchestrate multiple AI agents with shared context and automatic task handoffs.
npx skills add https://github.com/Castrozan/.dotfiles --skill crewai-castrozan
Please help me install this Agent Skill. Skill: crewai Source: https://github.com/Castrozan/.dotfiles/tree/main/agents/skills/crewai Command: npx skills add https://github.com/Castrozan/.dotfiles --skill crewai-castrozan
CrewAI provides a framework to orchestrate role-based AI agents that collaborate with shared context and automatic task handoff, enabling teams to work together more efficiently.
Install CrewAI via pip install crewai crewai-tools From crewai import Agent, Task, Crew, Process
researcher = Agent(role='Senior Researcher', goal='Find and analyze technical information', tools=[SerperDevTool()], verbose=True) writer = Agent(role='Content Writer', goal='Craft engaging content', verbose=True)
research = Task(description='Research {topic}', expected_output='5 key findings', agent=researcher)
article = Task(description='Write article on {topic}', expected_output='Markdown article', agent=writer, context=[research])
High-intent search queries and answers about installing and using this skill.
Multi-agent orchestration coordinates role-based AI agents by defining Agents, Tasks, and Crews in Python. It supports sequential and hierarchical processes, allowing agents to pass shared context and hand off tasks automatically.
Agents share context automatically by defining Tasks with context dependencies, where downstream tasks reference upstream outputs. This allows a researcher agent to pass findings directly to a writer agent within the same Crew.
Yes, you can build event-driven multi-agent workflows using Flows. Flows enable processes to start, listen for specific events, and route data dynamically, supporting event-driven orchestration alongside standard sequential or hierarchical Crews.
Yes, you need the crewai and crewai-tools Python packages to run a multi-agent crew. These dependencies provide the core libraries to define Agents, Tasks, Crews, and integrate external tools like search or APIs.
The best way to structure this workflow is defining a sequential Crew with a researcher Task and a writer Task. The writer task references the research output as context, enabling automatic task handoff and shared context within the Python crew.
qcmuuOrchestrate role-based CrewAI agent teams for sequential or hierarchical task delegation.
arsityOrchestrate CrewAI agents and flows with memory and observability.
SstoboDefine Convex-aware tools with createTool and wire them into agents.
Geralt1983Execute multi-agent pipelines through sequential prompt chaining and predefined workflows.
