01
It starts by itself
agent config changed. Runs in GitHub Actions or Claude Code routine.
- ↻ agent config changed
push touches agents.yaml or crew.py
Every agent passes the checklist.
Install the CrewAI design agent skill and make it a habit: follow https://gethabit.dev/skills/crewaiinc/design-agent.md
8K installs47 starsMIT licenceSource on GitHub ↗

CrewAI design agent
by crewaiinc · 8K installs
The skill
The crewai design agent skill helps developers design CrewAI agents: role, goal and backstory, LLM choice, tools, limits, planning, delegation, knowledge sources and guardrails.
Only when you ask.
The habit
Every agent passes the checklist.
↻ Runs by itselfagent config changed
When does it start? How does it check its work? What does it remember?
Stops after reviewing 10 agents per run.
Ask whether the work really needs more than one agent.
Make roles specific and goals outcome-focused with quality standards.
Confirm data tasks have tools and each agent has at most five.
Tune max_iter and set LLM and tool timeouts instead of relying on max_execution_time.
Move guardrails for critical output onto the Task.
Write each checklist gap and its fix to the notes.
It writes only in its own folder and asks you before anything goes out.
Read both lists. If the left one sounds like you, the CrewAI design agent skill helps.
The skill says too many tools confuse the agent and suggests three to five.
From the check-crewai-api skill, which wins when the two disagree.
Copy the prompt at the top into your agent. It installs the skill from crewaiinc/skills and sets up the habit. Or run npx skills add crewaiinc/skills --skill design-agent.
Yes. It is published under MIT. Habit is free too.
Install the CrewAI design agent skill and make it a habit: follow https://gethabit.dev/skills/crewaiinc/design-agent.md
All crewaiinc skills →More code skills →
Works in
Made by crewaiinc, published under MIT. Habit links to the original and never changes it. Is this your skill?