Content
77%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, actionable research-planning skill with excellent workflow sequencing, validation checkpoints, and clean progressive disclosure via verified reference files. Its main weakness is conciseness: the Hard Rules section duplicates guidance already present in the stepwise execution body.
Suggestions
Collapse the 'Hard Rules' section into a short non-negotiables list, removing rules that merely restate instructions already given inline in Steps 4.5, 5, and 7 (e.g., rules 11–13 on reference verification, 16/18/20 on dependency mapping, 22 on the dataset disclaimer) to cut duplication.
Move the per-config detail tables and the endpoint-formula examples into the existing reference files, keeping SKILL.md as a lean overview that points to them, which would tighten conciseness without losing actionability.
Consolidate the repeated 'mandatory section' assertions (rules 15, 20, 21, 23 and the Step 7 section list) into a single canonical checklist to reduce token cost.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient but the 23-item 'Hard Rules' section substantially re-states constraints already embedded in the execution steps (dependency consistency, reference verification, disclaimer, minimal-subset rules), adding noticeable padding rather than earning every token. | 3 / 5 |
Actionability | Provides concrete, executable guidance — config comparison tables, exact reference categories I1–I4, explicit dependency formulas, mandatory section lists, and the DESeq2-vs-limma data-type rule — with only minor gaps because bulk detail is delegated to reference files. | 4 / 5 |
Workflow Clarity | A clearly sequenced 7-step process ('always run in order') with explicit validation checkpoints — Step 5 dependency consistency check 'mandatory before output', Step 4.5 literature verification, redirect-and-stop out-of-scope handling, and the mandatory self-critical risk review — giving strong feedback-loop structure. | 5 / 5 |
Progressive Disclosure | Clear overview with well-signaled, one-level-deep references to references/*.md, all of which exist as real bundle files; content is appropriately split across SKILL.md and the reference modules for easy navigation. | 5 / 5 |
Total | 17 / 20 Passed |