skill

crewai — Claude Code skill: build CrewAI agent teams

Q: How do I install crewai in Claude?

Use the templates CLI: `npx claude-code-templates@latest --skill ai-research/crewai`. Or add it with `npx skills add davila7/claude-code-templates --skill crewai -a claude-code`.

Q: What do I need to run it?

Python 3.10+, the `crewai` package installed, and LLM API access. The skill itself is Python-only.

Designs multi-agent teams on CrewAI: roles, tasks, orchestration and process types. Useful once a single agent stops being enough.

Install into Claude Code · claude-code-templates All methods and platforms ↓

npx claude-code-templates@latest --skill ai-research/crewai

Open on GitHub

What crewai does

The skill turns Claude into an architect for CrewAI multi-agent systems. It covers agent definitions built from role, goal and backstory, task design with expected outputs, and chaining tasks so one feeds the next. Process types get their own section: sequential, hierarchical and parallel, along with memory and tool integration.

Templates come with it. The basic pattern splits config into config/agents.yaml and config/tasks.yaml, then a @CrewBase class with @agent and @task decorators assembles the crew. For harder cases there is a hierarchical process where a manager agent picks who handles what, and a planning mode that generates a step-by-step plan before the run and injects it into every task.

You need Python 3.10+, the crewai package and LLM API access. Everything here is Python, so a project in another language gets nothing from it. The skill also walks through common mistakes: vague roles, missing expected outputs, and crews that grow too large — past three to five agents the coordination overhead starts to outweigh the gain.

What it helps with

define agents with roles and goals instead of one generic prompt
chain tasks so each one consumes the previous output
pick between sequential and hierarchical processes
configure crew memory and wire up external tools
break a complex workflow into steps with planning

Example prompts for the agent

Build a crew of three agents: researcher, analyst and writer
Switch this crew to a hierarchical process with a manager
Turn on planning and show me the plan before the run

How to install crewai

1claude-code-templates

npx claude-code-templates@latest --skill ai-research/crewai

2Skills CLI (Vercel)

npx skills add davila7/claude-code-templates --skill crewai -a claude-code

3Manually: put the files in place

~/.claude/skills/crewai/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/crewai/

Manually: put the files in place

.agents/skills/crewai/SKILL.md

For all projects use ~/.agents/skills/

Manually: put the files in place

.cursor/skills/crewai/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install davila7/claude-code-templates crewai

2Manually: put the files in place

.github/skills/crewai/SKILL.md

Platform CLI

gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/crewai

Manually: put the files in place

.devin/skills/crewai/SKILL.md

The legacy .windsurf/skills/ path still works

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about crewai

How do I install crewai in Claude?

Use the templates CLI: npx claude-code-templates@latest --skill ai-research/crewai. Or add it with npx skills add davila7/claude-code-templates --skill crewai -a claude-code.

What does the crewai skill do?

It helps you design CrewAI multi-agent teams: agent roles and backstories, tasks with expected outputs, orchestration and process types. It also covers anti-patterns like vague roles.

What do I need to run it?

Python 3.10+, the crewai package installed, and LLM API access. The skill itself is Python-only.

multi-agent-orchestration qodex-ai Guides the architecture of multi-agent setups, delegating specialized roles and combining outputs. Helpful for multi-stage workflows across multiple LLMs. CX Codex +1 2,100installs check-crewai-api crewAIInc A skill that checks crewai code against 1.15.x instead of the 0.x syntax assistants still remember. Useful when an import or a parameter quietly does nothing. CX Codex +1 650installs agent-orchestration OrchestKit A skill of patterns for autonomous agent loops, multi-agent coordination and multi-scenario runs. Useful when you build agent pipelines and pick a framework. CX Codex +1 1,400installs design-agent crewAIInc Teaches your AI assistant how to configure CrewAI agents: define roles, set goals and backstories, assign tools, and adjust execution limits without token waste. CX Codex +1 8,300installs agent-designer Alireza Rezvani Turns written requirements into a multi-agent architecture, validated tool schemas and a read on your run logs. Worth running before you wire agents to a design you picked by gut. 682installs agent-workflow-designer Alireza Rezvani Picks a multi-agent workflow pattern and scaffolds the config before you write code, with handoff contracts and retry gates spelled out. 658installs

From the same set205

The whole set →

change-logger This hook writes each file mutation to a CSV: timestamp, tool, path, action and details. Handy when you are prepping a demo and need to show what the agent touched. 163installs git-bisect-helper Runs a git bisect session and marks commits good or bad from your test's exit code. Useful when you know something broke but not where. 163installs cloud-architect A Claude skill for designing infrastructure across AWS, Azure and GCP, with Terraform, CDK, cost work and security controls. 157installs timeline-compressor Compresses a real-world timeline into fast simulation cycles so you can test decisions and risks sooner. Useful before you commit to a plan. 146installs perplexity This skill wires Perplexity into Claude: web search, conversational answers, and deep research through an agent. Handy when you need current information rather than library docs. 140installs dependency-updater A skill that detects your project type, applies safe dependency updates on its own and asks before touching major versions. 136installs data-processing-ray-data A Claude skill that tells the agent how to process large datasets with Ray Data: streaming, GPU transforms, batch inference. 134installs svelte-debug Reads a Svelte or SvelteKit error and stack trace, finds whether the cause is compilation, runtime or config, and hands you a fix. 134installs