Content
56%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 contains genuinely actionable material — a parameter table, runnable commands, a validated workflow with fallback handling, and real reference files — but it is wrapped in heavy template boilerplate and self-referential filler that roughly doubles its length. Tightening it to the domain-specific sections (Usage, How It Works, References, Workflow, Error Handling) and moving meta-content out of SKILL.md would substantially improve it.
Suggestions
Cut the boilerplate and duplication: remove the verbatim description repeats in "When to Use"/"Key Features" and delete or split out generic sections (Key Features, Implementation Details, Lifecycle Status, Security Checklist, Evaluation Criteria, Response Template) into a separate file or remove them.
Fix navigation defects: replace forward-dangling "See `## Prerequisites` above" pointers (which appear before their targets), remove the hardcoded cd into "20260318/scientific-skills/Evidence Insight/journal-matchmaker", and ship or delete the missing requirements.txt.
Reconcile the CLI documentation: document the script's actual --file flag and correct the "--abstract \"abstract.txt\"" example so examples are copy-paste consistent with the parameter table.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The frontmatter description is repeated verbatim four times ("Analyzes academic paper abstracts to recommend optimal journals..." in the intro, When to Use, and Key Features), and generic boilerplate sections (Key Features, Implementation Details, Lifecycle Status, Security Checklist, Evaluation Criteria) add no skill-specific knowledge. A stale date ("Next Review Date: 2026-03-06") also pads the body, making this noticeably verbose rather than just loosely tighten-able. | 2 / 5 |
Actionability | The body gives concrete, mostly executable commands ("python scripts/main.py --abstract \"...\" --format json") whose flags match the actual argparse in scripts/main.py, plus a complete parameter table. Minor gaps keep it below 5: the script's "--file" flag is undocumented, the example "--abstract \"abstract.txt\"" contradicts the table's "abstract text" description, and requirements.txt is referenced but absent from the bundle. | 4 / 5 |
Workflow Clarity | The Workflow section sequences five steps with an early-stop validation ("stop early if the task would require unsupported assumptions"), a py_compile Quick Check, and an explicit failure/fallback path with stated blocking causes. It falls short of 5 because two competing process descriptions (the generic "Workflow" and the domain-specific "How It Works") dilute which sequence to follow, and the domain steps lack checkpoints. | 4 / 5 |
Progressive Disclosure | Bundle references are real, one level deep, and clearly listed ("references/journals.json - Journal database with impact factors and scopes"), but roughly a hundred lines of meta-content (Risk Assessment, Security Checklist, Evaluation Criteria, Response Template, Output Requirements) are inlined in SKILL.md where they belong in separate files. Forward-dangling pointers ("See `## Prerequisites` above" appearing before that section exists) further blur navigation, so structure is present but not well organized. | 3 / 5 |
Total | 13 / 20 Passed |