Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill's single action is concrete and actionable, but the body is marred by triple repetition, a hostile filler line, and a leftover TODO, with no verification step or organization. It communicates intent yet reads as unpolished and unprofessional.
Suggestions
Remove the profanity, the TODO comment, and the duplicate restatements; state the instruction once, cleanly.
Add a brief verification step for the batch read (e.g., 'List docs/adr/*.md first and confirm each was read in full') to raise workflow clarity.
Organize with a minimal heading (e.g., '## What to do') so the short body reads as intentionally structured.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short but padded: the core 'read every ADR' instruction is repeated three times and includes a profanity-laced exhortation ('DO NOT BE A LAZY CUNT. ACTUALLY DO THIS.') and a leftover TODO comment that earn no informational value. | 2 / 3 |
Actionability | For this single-purpose instruction skill, 'Read EVERY single ADR .md file in this project's docs/adr/ folder, start to finish' is a concrete, unambiguous, executable instruction with a specific path; absence of code is not penalized when guidance is actionable. | 3 / 3 |
Workflow Clarity | The single action is unambiguous, but this is a batch read operation with no verification step (e.g., listing/confirming all ADRs were read), and the redundant restatements obscure the sequence rather than clarifying it. | 2 / 3 |
Progressive Disclosure | The body is tiny and needs no external references, but it has no organized sections (only prose, a TODO comment, and filler), so it does not meet the 'well-organized sections' bar for a 3 on short skills. | 2 / 3 |
Total | 9 / 12 Passed |