crewai-multi-agent

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/llm-tools/crewai/SKILL.md

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, highly actionable reference with genuinely useful progressive disclosure into three real bundle files. Its main cost is token bloat from redundant parallel paths and conceptual padding Claude does not need, plus a couple of non-self-contained code examples.

Suggestions

Collapse the three parallel quick-start paths to one canonical path and move the full YAML project walkthrough into a reference file to cut roughly a third of the body.

Drop or shrink the 'CrewAI vs alternatives' comparison table and the repeated '50+ tools' claims — Claude already knows the LangChain/LangGraph landscape.

Make the Flows example self-contained (define analysis_crew/report_crew or note they are defined elsewhere) and replace the eval()-based CalculatorTool with a safer implementation so the examples are truly copy-paste ready.

Dimension	Reasoning	Score
Conciseness	The body is mostly tight code examples, but it presents three parallel quick-start paths (CLI, code-only, and a full YAML project walkthrough), a 'CrewAI vs alternatives' table that largely restates knowledge Claude already has, and repeats '50+ tools' three times — several sections could be trimmed or consolidated.	3 / 5
Actionability	Nearly all code is complete and copy-paste ready (agents, tasks, crews, process types, tools, YAML config, memory, LLM providers), but the Flows example references undefined variables (analysis_crew, report_crew) and the custom-tool example uses eval(), leaving minor gaps from fully executable coverage.	4 / 5
Workflow Clarity	The content follows a clear install → create → run progression with a logically ordered concept buildup, and the 'Common issues' section provides error-recovery patterns; no explicit validation checkpoints exist, though this library-usage skill involves no destructive or batch operations requiring them.	4 / 5
Progressive Disclosure	Three real, well-signaled, one-level-deep reference files (flows.md, tools.md, troubleshooting.md) are linked from a clear References section and inline at the point of need, but the ~490-line body still inlines material (the full YAML project walkthrough, the alternatives comparison table) that could be split out.	4 / 5
	Total	15 / 20 Passed

Description

78%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with an explicit, multi-clause 'Use when...' trigger section and concrete capability statements. Its main weaknesses are the missing primary trigger terms ('CrewAI', 'crew') and a second-person construction that costs it specificity.

Suggestions

Add the framework's own natural trigger terms ('CrewAI', 'crew', 'flows') so users who name the library directly will match the skill.

Rewrite the second-person clause 'when you need role-based agent collaboration' in third person or imperative form (e.g., 'Use when role-based agent collaboration is needed') to satisfy the voice guideline.

Replace the marketing-flavored closer 'Built without LangChain dependencies for lean, fast execution' with a concrete distinguishing capability to sharpen both specificity and distinctiveness.

Dimension	Reasoning	Score
Specificity	Concrete capabilities are named ('building teams of specialized agents working together', 'role-based agent collaboration with memory', 'sequential/hierarchical execution'), matching the anchor for several specific actions with minor gaps, but the second-person phrasing 'when you need role-based agent collaboration' triggers the mandated 1-point specificity penalty.	3 / 5
Completeness	The description clearly answers 'what' ('Multi-agent orchestration framework for autonomous AI collaboration') and explicitly answers 'when' with multiple concrete trigger clauses ('Use when building teams of specialized agents... when you need role-based agent collaboration... or for production workflows requiring sequential/hierarchical execution').	5 / 5
Trigger Term Quality	Natural phrases like 'multi-agent', 'teams of specialized agents', 'role-based', 'agent collaboration', and 'workflows' are present, but the primary natural trigger terms 'CrewAI', 'crew', and 'flows' are absent, leaving a few common terms missing.	4 / 5
Distinctiveness Conflict Risk	A clear multi-agent orchestration niche with 'Built without LangChain dependencies' actively disambiguating from LangChain/LangGraph skills, though it still overlaps somewhat with other agent-framework skills.	4 / 5
	Total	16 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (508 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/crewai/SKILL.md
Commit: 4082a2e

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.