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.
The body is a well-structured, actionable research-design framework with clear sequencing, an explicit dependency-check validation loop, and excellent progressive disclosure via verified reference files. Its main weakness is redundancy between the Hard Rules and the step/output sections.
Suggestions
Consolidate the 23 'Hard Rules' with the Step 5 dependency check and Step 7 mandatory-section list — many rules (15, 17, 20, 22, 23) restate those requirements, adding tokens without new information.
Move any repeated guidance (e.g., the Dataset Disclaimer mandate, which appears in Step 7-D and rule 22) to a single authoritative location and reference it once.
Tighten the 'Execution — 7 Steps' intro and the Hard Rules list by keeping only rules that add constraints not already implied by the step structure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly directive and free of generic background padding, but the 23-item 'Hard Rules' section substantially restates mandates already specified in Step 5 and the Step 7 output sections (e.g., rules 15, 17, 20, 22, 23 duplicate earlier requirements), which could be consolidated. | 3 / 5 |
Actionability | Provides concrete, specific guidance — a 7-step ordered process, study-pattern and config tables, mandatory A–J output sections, and exact rules (DESeq2 for raw counts, limma for non-count matrices; DOI/PMID/PMCID required for references) — with only minor gaps since most granular detail is delegated to the eight reference files. | 4 / 5 |
Workflow Clarity | Clear ordered 7-step sequence with an explicit validation checkpoint (Step 5 dependency consistency check, an 8-item checklist) and a feedback loop ('If any dependency inconsistency is found, revise the plan before outputting'), plus an out-of-scope redirect and validation-tier separation. | 5 / 5 |
Progressive Disclosure | SKILL.md is a well-organized overview that points to eight clearly-signaled, one-level-deep reference files via markdown links; all referenced paths (study-patterns.md, workload-configurations.md, literature-retrieval-and-citation.md, workflow-step-template.md, analysis-modules.md, method-library.md, figure-deliverable-plan.md, validation-evidence-hierarchy.md) exist as real files. | 5 / 5 |
Total | 17 / 20 Passed |