agent-swarm
Agent skill for swarm - invoke with $agent-swarm
40
7%
Does it follow best practices?
Impact
100%
2.27xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agents/skills/agent-swarm/SKILL.mdQuality
Discovery
0%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 an extremely weak description that fails on every dimension. It provides no information about what the skill does, when it should be used, or what distinguishes it from other skills. It reads more like a label than a functional description that Claude could use for skill selection.
Suggestions
Describe the concrete actions this skill performs (e.g., 'Coordinates multiple agents to perform parallel tasks, distributes workloads, and aggregates results').
Add an explicit 'Use when...' clause with natural trigger terms that describe scenarios where this skill should be selected (e.g., 'Use when the user needs to run multiple tasks in parallel, coordinate agent workflows, or distribute work across agents').
Remove the invocation command ('invoke with $agent-swarm') from the description and replace it with capability-focused language that helps Claude distinguish this skill from others.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description contains no concrete actions whatsoever. 'Agent skill for swarm' is entirely vague and abstract, providing no information about what the skill actually does. | 1 / 3 |
Completeness | The description fails to answer both 'what does this do' and 'when should Claude use it'. There is no explanation of capabilities and no 'Use when...' clause or equivalent trigger guidance. | 1 / 3 |
Trigger Term Quality | The only keyword is 'swarm', which is technical jargon without context. There are no natural user-facing trigger terms that someone would use when needing this skill. '$agent-swarm' is an invocation command, not a trigger term. | 1 / 3 |
Distinctiveness Conflict Risk | The description is so vague that it provides no distinguishing characteristics. 'Agent skill for swarm' could mean virtually anything and offers no clear niche or distinct triggers to differentiate it from other skills. | 1 / 3 |
Total | 4 / 12 Passed |
Implementation
14%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads more like a persona description or marketing document than actionable instructions. While it includes useful MCP tool call examples, the majority of the content is verbose descriptions of concepts, capabilities, and quality standards that don't help Claude execute specific tasks. It lacks concrete workflows, validation steps, error handling guidance, and progressive disclosure structure.
Suggestions
Remove the descriptive sections (core responsibilities, agent types descriptions, quality standards, topology descriptions) and replace with a concise quick-start workflow showing a complete swarm deployment from init to task completion with actual tool calls and expected responses.
Add explicit validation checkpoints: after swarm_init check status, after agent_spawn verify agents are ready, after task_orchestrate monitor for completion/failure with concrete decision logic for error recovery.
Replace the abstract 6-step 'orchestration approach' with concrete decision trees: e.g., 'If task requires parallel independent subtasks → use mesh topology; if sequential pipeline → use ring topology' with specific tool call sequences for each.
Extract detailed topology selection guides and agent capability matrices into separate referenced files, keeping SKILL.md as a lean overview with the most common workflow pattern.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose with extensive explanations of concepts Claude already knows (what topologies are, what agent types do, generic quality standards). The bullet-point lists of agent types and topologies with descriptions are padding that doesn't add actionable value. The 'Your core responsibilities' and 'Quality standards' sections describe rather than instruct. | 1 / 3 |
Actionability | The JavaScript code block provides concrete MCP tool calls with parameter examples, which is useful. However, the code is illustrative rather than executable in a real workflow—there's no guidance on how to chain these calls together, handle responses, or make decisions based on outputs. Much of the content describes capabilities rather than providing step-by-step instructions. | 2 / 3 |
Workflow Clarity | The 6-step 'orchestration approach' is vague and abstract (e.g., 'Track swarm efficiency and agent utilization') with no validation checkpoints, error recovery steps, or concrete decision criteria. There's no guidance on what to do when agents fail, how to verify swarm health before proceeding, or when to scale vs. destroy. For a multi-step orchestration process, this lacks the necessary specificity. | 1 / 3 |
Progressive Disclosure | The content is a monolithic wall of text with no references to external files and no clear separation between quick-start and advanced content. Everything is dumped into a single file with no navigation structure or links to detailed documentation for specific topics like topology selection or error handling. | 1 / 3 |
Total | 5 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- ruvnet/claude-flow
- Commit
01070ed
- Reviewed
Table of Contents
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.