Content
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is lean and highly actionable, with executable curl/python recipes for each common HN intent, a clear output format, and a sensible fallback for paywalled pages. It is well-organized as a single file with no unnecessary explanation of concepts Claude already knows.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean, assumes Claude's competence, and avoids explaining generic concepts; even the opening paragraph earns its place by stating a non-obvious gotcha ("do not scrape" hckrnews, use the JSON APIs instead) rather than filler, matching the level-3 anchor. | 3 / 3 |
Actionability | Each section provides fully executable, copy-paste-ready curl + python3 one-liners with clearly marked parameter placeholders (YOUR+QUERY, OBJECT_ID, ARTICLE_URL), matching the level-3 anchor for executable, specific examples. | 3 / 3 |
Workflow Clarity | Distinct intents are sequenced into clearly numbered sections (1 top stories, 2 search, 3 story+comments, 4 article) with a defined output spec and a recovery path for paywalled/JS-heavy pages; the task is read-only so no validation checkpoints are required, satisfying the simple-skill allowance for level 3. | 3 / 3 |
Progressive Disclosure | The skill is a self-contained single file (no bundle files present) with well-organized section headers and no nested reference chains; the inline code is core action content rather than material that should be split out, matching the level-3 allowance for simple skills with well-organized sections. | 3 / 3 |
Total | 12 / 12 Passed |