Content
38%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.
The body reads like a generic skill template stretched over a citation-formatter use case: duplicated trigger sections, placeholder code blocks, contradictory imports, and a large fraction of dead references — none of the cited bundle files (scripts/main.py, references/guide.md, references/examples/, references/api-docs/) actually exist. The genuinely useful material (Quick Start API calls, CLI example, best practices) is buried in filler. It is not hopeless — there is a real workflow skeleton and concrete API guidance — but the file needs substantial de-duplication and either the missing bundle files or removal of the references to them.
Suggestions
Remove the duplicated 'When to Use' / 'When to Use This Skill' sections, the description re-pasted into 'Key Features', and the generic 'Implementation Details' / 'Quality Checklist' boilerplate; keep one concise trigger section.
Replace the four 'result = tool.execute(data)' placeholder blocks in Core Capabilities with the real executable code (or delete the sections), and fix the contradictory Quick Start imports (scripts.main vs scripts.citation_formatter).
Either ship the referenced bundle files (scripts/main.py, references/guide.md, references/examples/, references/api-docs/) or delete all references to them, since every cited path is currently dead.
Rewrite the generic Workflow steps as a citation-specific sequence (validate input references → convert/format → run validation → report incomplete references) with an explicit validate-and-fix loop for batch operations.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is noticeably verbose and padded: the trigger guidance appears in two duplicated sections ("When to Use" and "When to Use This Skill"), the frontmatter description is re-pasted verbatim inside "Key Features" ("Scope-focused workflow aligned to: Use when formatting references for journal submission..."), and generic template boilerplate fills "Implementation Details" and "Quality Checklist" ("Clear understanding of your objectives"). It does not reach anchor 1 because it never explains concepts Claude already knows — the waste is duplication and filler, not tutorials. | 2 / 5 |
Actionability | The Quick Start gives concrete, plausible API calls (formatter.convert_style(from_style="APA 7th", to_style="Vancouver", include_doi=True)) and a CLI line, but four "Core Capabilities" blocks are pure placeholders ("result = tool.execute(data)"), the Quick Start contains two contradictory imports (scripts.main vs scripts.citation_formatter), and every runnable command targets scripts/main.py, which does not exist in the bundle. This mixes executable-looking guidance with pseudocode and dead paths, matching anchor 3; it is above anchor 2 because specific method signatures and parameters are given. | 3 / 5 |
Workflow Clarity | A numbered "Workflow" and "Example run plan" exist and a py_compile "Quick Check" provides one validation checkpoint, but the steps are generic template language ("Confirm the user objective... choose the packaged workflow") rather than a citation-specific sequence, fragmented across redundant sections (Workflow, Example run plan, Audit-Ready Commands, Response Template). Batch operations (batch_format) lack a real validate→fix→retry feedback loop, capping this at 3 per the rubric. | 3 / 5 |
Progressive Disclosure | Scored against the actual bundle structure: no references/, scripts/, or assets/ directories exist, so the "References" section (references/guide.md, references/examples/, references/api-docs/) and every scripts/main.py invocation point to files that are not present — navigation to detail is impossible. Content that belongs in separate files (API usage, examples) is inlined in the body instead. This matches anchor 2 (content that clearly belongs in separate files is inlined / broken reference structure); it is above anchor 1 because section headers do provide navigable structure within the body. | 2 / 5 |
Total | 10 / 20 Passed |