PlugWiki

design-task skill (crewAIInc): write and configure CrewAI tasks

crewAI's own skill that guides a coding agent on writing and configuring CrewAI tasks, structured output and guardrails.

Installs
8.4K
on skills.sh
Stars
48
on GitHub
Files
1
besides SKILL.md
Updated
Oct 2026
last repo commit

design-task makes your coding agent follow CrewAI's rules for writing tasks: a clear description, a concrete expected_output, one purpose per task, and the right context between tasks. It also covers structured output, guardrails, async tasks and human review.

It loads when you create, configure or debug a CrewAI task. It edits your task code and YAML. SKILL.md says check-crewai-api is the reference for exact API forms and wins when they disagree.

Good for

  • Python developers building agents or crews with the CrewAI framework
  • Developers who need a task to return a validated Pydantic object
  • People debugging tasks that loop, return vague text or ignore earlier results

Install design-task

Claude Code

  1. 1

    Copy this prompt.

    Install the design-task skill in Claude Code. Run this command:

/plugin marketplace add crewAIInc/skills
/plugin install crewai-skills@crewai-plugins

Then check that it loaded and tell me how to use it.
  2. 2

    Paste it into Claude Code. It shows the change and asks before running it.

Pick one route, not both, or you get every skill twice. The README documents the plugin route for Claude Code. The plugin installs all 10 skills in the crewai-skills plugin; the skills CLI route installs design-task, check-crewai-api. design-task says to follow check-crewai-api where they disagree. The other CrewAI skills are optional hand-offs.

From the project README. Source.

What's in the skill

When it loads

CrewAI task design and configuration. Use when creating, configuring, or debugging crewAI tasks — writing descriptions and expected_output, setting up task dependencies with context, configuring output formats (output_pydantic, output_json, output_file), using guardrails for validation, enabling human_input, async execution, markdown formatting, or debugging task execution issues.

Sections in SKILL.md

  1. 1. Anatomy of an Effective Task
  2. Description — The Instructions
  3. Expected Output — The Success Criteria
  4. 2. The Single Purpose Principle
  5. 3. Task Configuration Reference
  6. Essential Parameters
  7. Task Dependencies with context
  8. Structured Output
  9. File Output
  10. Async Execution
  11. Human Review
  12. Markdown Formatting
  13. Callbacks
  14. 4. Task Guardrails — Quality Control
  15. Function-Based Guardrails
  16. LLM-Based Guardrails

Key facts

Author
crewAI (the crewAIInc organisation); the plugin manifest names crewAI as author and the repo is the vendor's own
License
MIT (LICENSE file and README agree)
Installs on skills.sh
8,376 as of 2026-11-02
GitHub stars
48 for the whole crewAIInc/skills repo, 2026-11-02
What is inside
2 files: SKILL.md and references/structured-output.md. It bundles no scripts.
Last change to this skill
2026-10-06
Part of
The crewai-skills plugin (10 skills) in the crewai-plugins marketplace
Checked against
crewai 1.15.23 on 2026-10-01, per the SKILL.md

What it includes

  • Description for the instructions, expected_output for the result

    Both are required: Task(description=...) alone raises a validation error on expected_output.

    Write · Anatomy of a task

  • Split god tasks into focused ones

    Shows a task with four objectives and the same work split into separate tasks.

    Write · Single purpose principle

  • Parameters, context, structured output, files, async, human review

    Covers context=[...] rules, output_pydantic, output_json and output_file, async_execution rules and Task(human_input=True).

    Write · Configuration reference

  • Function-based and LLM-based checks, chainable

    Explains guardrails that validate and retry a task's output, and how to chain several.

    Write · Task guardrails

  • Define tasks in tasks.yaml and wire them in crew.py

    The recommended setup, with method names matching YAML keys.

    Write · YAML configuration

  • Diamond dependencies and ConditionalTask

    Shows context flow between tasks and when a ConditionalTask runs.

    Write · Dependencies and conditional tasks

  • Common task design errors and a closing review list

    Sections 9 and 10.

    Read · Mistakes and checklist

  • Structured output at LLM, Agent, Task and Crew level

    Shows which keyword each level takes, such as response_model on LLM.call() and response_format on Agent.kickoff().

    Read · references/structured-output.md

From the official docs

Permissions

Low risk

The agent edits your CrewAI task code and config, and can run tasks that call your model provider with your own API key. Review the diff and cap provider spending before you run what it writes.

What it can do

  • Read your data
  • Create and change things
  • Delete things
  • Read-only mode

Watch out for

  • It edits your project files

    It changes task definitions and config. Work in a git branch and read the diff. github.com

  • human_input=True fails where there is no terminal

    SKILL.md says it raises EOFError in CI or a server after the agent's retries, so you pay for the extra LLM calls. Use Flow @human_feedback there. github.com

  • The plugin route installs all 10 skills

    The crewai-skills plugin manifest lists 10 skills and holds no hooks, agents or MCP config in the git tree. Use the skills CLI route if you want only the skills this page names. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details
Talks toYour model provider
Sign-in and keysProvider keys come from your environment.
Scopes

  • Edits task code and YAML config in your project

  • Can run crews and tasks, which call your model provider

  • SKILL.md shows human_input=True, which waits for typed input in the terminal

Questions

What does the design-task skill do?

It guides your coding agent on writing CrewAI tasks: description and expected_output, one purpose per task, context between tasks, structured output, guardrails, async tasks and human review.

How do I install design-task?

Use the install box at the top of this page. The README documents a Claude Code plugin with all 10 CrewAI skills. The skills CLI route in the box adds check-crewai-api, which design-task says is the reference where the two disagree.

Does it need other skills?

It names check-crewai-api as the reference for exact API forms and says it wins on disagreements, so the skills CLI route installs it. build-flow, named for the Flow @human_feedback alternative, is an optional hand-off.

Does design-task need an API key?

The skill itself does not. Running crews and tasks needs a model provider key, such as OPENAI_API_KEY for the default model.

Is design-task safe?

It edits your project files and may run tasks that call your model provider. Use a branch, read the diff and cap your provider spending.

How do I get a task to return a Pydantic object?

SKILL.md says to set output_pydantic in Python, not in tasks.yaml. Task(response_format=...) is silently ignored and Task(response_model=...) leaves .pydantic as None.

Is this the official CrewAI skill?

Yes, it is in the crewAIInc organisation's own repo, and the plugin manifest names crewAI as author.

Does it work in the Claude app or Codex?

The README documents Claude Code only. The skill edits project files, so there is no Claude app upload option here. The box lists the Codex skills folder as a generic route.

Information

updated Nov 2, 2026

Repository
crewaiinc/skills
Folder
skills/design-task

crewaiinc/skillsskills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 4
  1. README.md · github.com
  2. marketplace.json · github.com
  3. plugin.json · github.com
  4. SKILL.md · github.com

Updated Nov 2, 2026.