Content
71%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.
A highly actionable, well-structured skill body with executable commands, ready-made SQL examples, and meaningful safety checkpoints for destructive writes. Its weaknesses are token efficiency and file structure: a ~220-line inline table catalog (partly redundant with the live schema command the skill itself designates as authoritative) and a repeated systemconfig gating explanation.
Suggestions
Move the Core Tables catalog (the per-table Purpose/Useful queries/Write boundary/Columns entries) into a references/ file such as references/core-tables.md, keeping only the most-frequently-used tables (downloadhistory, subscribe, transferhistory, site) inline with a pointer to the full catalog.
Explain the systemconfig gate once (e.g., in Built-in Safety) and reference it briefly from the other sections instead of restating the rule in Scope And Boundaries, Commands, and Built-in Safety.
Add a post-write verification step to the Workflow (e.g., re-run the query after a write to confirm the intended rows changed and the affected row count matched expectations), which would also raise workflow clarity.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The command, workflow, and safety sections are tight, but the ~220-line Core Tables catalog inlines full column lists even though the document itself says 'Always treat the live schema <table> result as authoritative', and the systemconfig gating rule is explained three separate times (Scope And Boundaries, Commands, Built-in Safety). This is 'mostly efficient but includes some unnecessary explanation or could be tightened' — above the 2 anchor's pervasive padding but below the 4 anchor's minor trimmings. | 3 / 5 |
Actionability | Every command is copy-paste ready ('python scripts/mp-db.py tables', 'schema downloadhistory', 'query --file /path/to/query.sql', the --show-secrets example) and the Common Queries section provides nine complete SQL examples covering the common cases, plus a dialect-comparison table and troubleshooting entries. This matches the 5 anchor's 'fully executable; copy-paste ready code or commands; specific examples cover the common cases'. | 5 / 5 |
Workflow Clarity | The Workflow is a clear five-step sequence (prefer existing tools → tables → schema → narrow SELECT with LIMIT → write and report affected row count) with real checkpoints for destructive operations: confirm first, suggest a backup, require WHERE clauses, and 'verify the managed API can read the repaired value' for systemconfig repairs. This satisfies the database-operations feedback-loop expectation enough to avoid the cap-at-3, but a general post-write re-verification step (re-query to confirm the change) is missing, so it sits at the 4 anchor's 'most checkpoints present; minor validation gaps' rather than 5. | 4 / 5 |
Progressive Disclosure | Section headers are clear and the single bundle reference (scripts/mp-db.py) is real, correctly pathed, and one level deep, but the Core Tables catalog is roughly 200 lines of schema-reference material inlined in SKILL.md — exactly the anchor-3 pattern of '[200 lines of API reference that could be in a separate file]'. It is better organized than anchor 3's example and has no nesting problems, but the bulk reference content that should live in references/ keeps it below anchor 4's 'most content is appropriately placed'. | 3 / 5 |
Total | 15 / 20 Passed |