cursor-agent-supervisor
Offloading tasks with a well-defined scope to sub-agents, for instance to use a sub-agent to implement a set of specs. Use this skill whenever a task should not need a broad knowledge of the whole project
74
62%
Does it follow best practices?
Impact
91%
1.59xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./cursor-agent-supervisor/SKILL.mdQuality
Discovery
25%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is too abstract and relies on internal jargon rather than concrete, user-facing language. It fails to enumerate specific capabilities (e.g., spawning sub-agents, defining task boundaries, collecting results) and its trigger condition is defined negatively ('should not need broad knowledge') rather than with clear positive triggers. The description would benefit significantly from concrete actions and natural user-facing keywords.
Suggestions
List specific concrete actions such as 'Spawns sub-agents to implement specifications, run isolated tests, or complete self-contained coding tasks in parallel'.
Replace the vague 'when' clause with explicit trigger terms like 'Use when the user asks to delegate work, break a task into subtasks, implement specs independently, or run isolated sub-tasks that don't require full project context'.
Add natural keywords users would say, such as 'delegate', 'subtask', 'parallel', 'isolated task', 'break up work', 'sub-agent', 'spawn agent'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague language like 'offloading tasks' and 'well-defined scope' without listing concrete actions. The example of 'implement a set of specs' is a single, loosely defined example rather than specific capabilities. | 1 / 3 |
Completeness | It partially answers 'what' (offloading tasks to sub-agents) and has a 'when' clause ('whenever a task should not need a broad knowledge of the whole project'), but the 'when' is vague and defined by negation rather than explicit positive triggers. | 2 / 3 |
Trigger Term Quality | The description lacks natural keywords users would say. Terms like 'sub-agents', 'offloading tasks', and 'well-defined scope' are internal/technical jargon. Users would more likely say 'delegate', 'parallel tasks', 'break up work', or 'implement specs'. | 1 / 3 |
Distinctiveness Conflict Risk | The concept of 'sub-agents' provides some distinctiveness, but 'offloading tasks with a well-defined scope' is broad enough to overlap with many task delegation or workflow management skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a high-quality skill that efficiently teaches Claude how to supervise sub-agents. It provides concrete commands, a useful task template, clear workflow sequencing with validation steps, and error recovery guidance—all without unnecessary verbosity. The content respects Claude's intelligence while providing the non-obvious, tool-specific details needed to execute effectively.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what sub-agents are conceptually or waste tokens on things Claude already knows. Every section provides actionable, non-obvious information like the timeout requirement, model selection trick, and the task template. | 3 / 3 |
Actionability | Provides concrete, executable commands for creating and resuming sub-agent conversations, includes the specific timeout value needed, gives a copy-paste task description template, and lists explicit verification steps. The model discovery trick (`--model unknown-model`) is a practical, specific tip. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced: setup (prepare the environment) → give instructions (with template) → wait for completion → verify results. The 'After Sub-agent Completes' section provides explicit validation checkpoints, and 'When Things Go Wrong' provides error recovery feedback loops for multiple failure modes. | 3 / 3 |
Progressive Disclosure | For a standalone skill with no bundle files, the content is well-organized into logical sections (Model Selection, Invocation, Setup, Instructions, After Completion, Error Handling) that are appropriately sized. No section is bloated, and the structure supports easy scanning and discovery without needing external files. | 3 / 3 |
Total | 12 / 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
- YPares/agent-skills
- Commit
d2b7dff
- 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.