Content
63%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 review-style skill body with a clean Check/Fix/Explain/Code Review sequence and exemplary one-level progressive disclosure to references/rule.md. Its weaknesses are the duplicated 'why it matters' explanation of a concept Claude already knows, and the lack of any executable command or example inline — the hash-generation command and markup examples all live in the reference file.
Suggestions
Trim or remove the opening 'CDN compromise' paragraph — it explains a concept Claude already knows and is duplicated in references/rule.md's 'Why It Matters' section.
Inline the one-line hash-generation command (e.g. `curl -s <url> | openssl dgst -sha384 -binary | openssl base64 -A`) in the Fix section so it is executable without opening the reference.
Add a short verify step after Fix (re-check the rendered markup / re-inspect shared components) mirroring rule.md's Verification section, closing the workflow loop in the body itself.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The opening paragraph ('When you load JavaScript from a CDN, you're trusting that CDN completely — if it's compromised, attackers can serve malicious JavaScript to all your users...') explains a concept Claude already knows, is duplicated nearly verbatim in references/rule.md's 'Why It Matters', and is repeated a third time by the Quick Reference bullet 'SRI blocks execution if the file hash doesn't match, preventing CDN compromise attacks'. Not 2 because the rest of the body is tight; not 4 because the same rationale appears in three places. | 3 / 5 |
Actionability | Concrete attribute-level guidance is present ('Add integrity="sha384-..." to <script> and <link> tags', 'Always pair integrity with crossorigin="anonymous"', 'Find all external <script> and <link rel=stylesheet> tags... that load from CDNs and don't have integrity attributes'), but the body contains no executable code or command — 'Generate hashes with openssl or online tools' defers the actual command to references/rule.md. Not 4 because there is no inline code block or command; not 2 because the attribute guidance is specific and directly usable. | 3 / 5 |
Workflow Clarity | The Check → Fix → Explain → Code Review sections give an unambiguous sequence, with a concrete discovery step ('Find all external <script> and <link rel=stylesheet> tags in this HTML that load from CDNs and don't have integrity attributes. List each one.') and a targeted code-review step ('Flag exact elements, attributes, and routes where the rendered HTML violates the rule'). Not 5 because there is no explicit verify-after-fix step in the body — the verification steps live only in references/rule.md and are not linked into the workflow. | 4 / 5 |
Progressive Disclosure | The body is a lean, well-sectioned overview with a clearly signaled one-level-deep reference: 'For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`'. The referenced file exists, holds all code examples and detail, and does not nest further references, so navigation is easy and the split is appropriate. | 5 / 5 |
Total | 15 / 20 Passed |