jbvc/crewai
Expert in CrewAI - the leading role-based multi-agent framework used by 60% of Fortune 500 companies. Covers agent design with roles and goals, task definition, crew orchestration, process types (sequential, hierarchical, parallel), memory systems, and flows for complex workflows. Essential for building collaborative AI agent teams. Use when: crewai, multi-agent team, agent roles, crew of agents, role-based agents.
73
73%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
85%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly identifies the CrewAI framework, lists specific capabilities, and includes an explicit 'Use when' clause. The main weakness is the marketing fluff ('leading role-based multi-agent framework used by 60% of Fortune 500 companies') which adds no selection value, and the trigger terms could be more natural and comprehensive. The trigger terms read more like keyword tags than phrases users would naturally say.
Suggestions
Remove the marketing claim '60% of Fortune 500 companies' as it adds no value for skill selection and wastes space.
Expand trigger terms to include more natural user phrasings like 'build AI agent team', 'agent collaboration', 'orchestrate multiple agents', 'CrewAI project' rather than just keyword-style tags.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: agent design with roles and goals, task definition, crew orchestration, process types (sequential, hierarchical, parallel), memory systems, and flows for complex workflows. | 3 / 3 |
Completeness | Clearly answers both 'what' (agent design, task definition, crew orchestration, process types, memory systems, flows) and 'when' with an explicit 'Use when:' clause listing trigger terms. | 3 / 3 |
Trigger Term Quality | The 'Use when' clause includes relevant terms like 'crewai', 'multi-agent team', 'agent roles', 'crew of agents', 'role-based agents', but these are somewhat narrow and miss natural user phrasings like 'build AI agents', 'agent collaboration', 'CrewAI tutorial', or 'orchestrate agents'. The terms feel more like keyword tags than natural language a user would say. | 2 / 3 |
Distinctiveness Conflict Risk | Clearly scoped to CrewAI specifically as a named framework, with distinct triggers like 'crewai', 'crew of agents', and 'role-based agents' that are unlikely to conflict with other agent framework skills (e.g., LangGraph, AutoGen). | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a solid structural overview of CrewAI with useful patterns and anti-patterns, but suffers from incomplete/truncated code examples that undermine actionability. The content is moderately concise but could be tightened, particularly in the YAML configuration example. Adding validation steps, error handling guidance, and fixing the truncated code would significantly improve quality.
Suggestions
Fix the truncated code in the Basic Crew example (the `writing_task` method is cut off mid-line) and ensure all examples are fully executable with necessary imports and variable definitions.
Add validation/debugging guidance: how to verify agents are working correctly, common error messages and fixes, and how to inspect intermediate task outputs.
Extract the lengthy YAML configuration example into a referenced file (e.g., YAML_CONFIG.md) and keep only a minimal example inline to improve progressive disclosure.
Remove undefined variable references in hierarchical and planning examples by either defining them inline or showing complete minimal working examples.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary verbosity in agent backstories and descriptions that could be trimmed. The YAML config example is quite long, and some explanatory comments are redundant. However, it mostly avoids explaining concepts Claude already knows. | 2 / 3 |
Actionability | The code examples are mostly concrete but the first (and largest) example is truncated mid-line (`config=self.tasks_config['research_task'])` cuts off), making it non-executable. The YAML-in-Python pattern is also unusual and may confuse execution. The hierarchical and planning examples reference undefined variables (research_task, analysis_task, etc.). | 2 / 3 |
Workflow Clarity | The patterns are presented in a logical progression (basic → hierarchical → planning), and the planning feature shows a numbered sequence. However, there are no validation checkpoints, no error handling guidance, no feedback loops for when agents fail or produce poor output, and no guidance on debugging crew execution. | 2 / 3 |
Progressive Disclosure | The content is reasonably structured with clear sections (Patterns, Anti-Patterns, Limitations, Related Skills), but it's somewhat monolithic. The YAML config example is very long and could be split into a separate reference file. The 'Related Skills' section mentions other skills but doesn't link to them. No references to external detailed docs. | 2 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
Reviewed
Table of Contents