Content
76%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 well-structured, highly actionable CLI skill body with executable examples and clean sectioning. Its main weakness is workflow clarity for the batch-scraping use case it promotes: validation and retry feedback loops are only mentioned abstractly rather than prescribed.
Suggestions
Add a concrete validation/retry loop for batch scraping, e.g. check each output file is non-empty and re-run failed URLs with the browser `fetch` fetcher before reporting results.
Remove or relativize the hardcoded wrapper path (`C:/Users/羽裳/.codex/...`) so the reference resolves regardless of install location.
Trim the opening paragraph that re-states the frontmatter description to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean, assumes Claude's competence, and avoids explaining basic concepts, but the opening paragraph re-states the frontmatter and a few transitional sentences (e.g. "This skill is CLI-first") could be trimmed. | 4 / 5 |
Actionability | Provides copy-paste-ready, executable commands across the common cases — `scrapling extract get "https://example.com" out.md`, `--css-selector`, browser `fetch` mode, install extras, and MCP launch — covering the typical workflows specifically. | 5 / 5 |
Workflow Clarity | A rough sequence exists (prerequisite check → install → extract), but the skill explicitly endorses batch URL scraping and only vaguely advises handling "timeouts, retries, and failures" with no validate→fix→retry checkpoint, so the batch-operation cap applies. | 3 / 5 |
Progressive Disclosure | Content is organized into clear, well-signaled sections with a single one-level reference to the bundled `scripts/scrapling.ps1` wrapper; the main gap is the hardcoded user-specific absolute path to that wrapper, which is not portable. | 4 / 5 |
Total | 16 / 20 Passed |