Content
32%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is generic agent-persona boilerplate: capability labels, work-style attitudes, and a vague output format with no concrete guidance, commands, or examples. It tells Claude how to behave rather than what to do, so it would not meaningfully change behavior on any specific task.
Suggestions
Replace the attitude-based 'Instructions' with a concrete task workflow (e.g. scope the task, plan, execute in checkpoints, verify against requirements) with explicit validation steps.
Add executable specifics for at least one or two of the named use cases — a sample analysis structure, a report template, or a verification command — instead of 'Code examples where applicable'.
Delete the duplicate embedded '---' frontmatter block at the top of the body and the filler closing line; fold any real signal (long-running, 30+ minute tasks) into the description instead.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and does not explain concepts Claude already knows, but sections like 'Remember: You have plenty of time to do thorough, high-quality work!' and the attitude list are filler that earns no tokens. This matches anchor 3 (mostly efficient, some unnecessary content) rather than 4, where every section would carry actionable weight. | 3 / 5 |
Actionability | There is no code, command, or concrete instruction anywhere — bullets like 'Complex Analysis: Deep dive into codebases, documentation, and systems' and 'Take Your Time: Don't rush - quality over speed' only describe attitudes. This squarely matches anchor 1 (entirely vague or abstract; only describes rather than instructs). | 1 / 5 |
Workflow Clarity | The 'Instructions' section is numbered but lists dispositions ('Be Thorough', 'Document Everything') rather than task steps, and there are no validation checkpoints or feedback loops. This matches anchor 2 (rough sequence present but steps poorly defined); not 1 because a numbered list and an explicit output-format section do exist. | 2 / 5 |
Progressive Disclosure | The skill is under 50 lines with no external references needed, and section headers (Capabilities, Instructions, Output Format, Example Use Cases) are clearly organized. It falls short of 5 because the body opens with a stray duplicate '---' frontmatter block, a structural defect that confuses navigation. | 4 / 5 |
Total | 10 / 20 Passed |