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 dense, well-structured pipeline skill with concrete tooling and clear phase sequencing, weakened by redundancy between its summary and body and by references to bundle files that are not present.
Suggestions
Provide the missing bundle files (phase-procedures.md, tool-reference.md, reference-data.md, report-template.md, test_spatial_omics.py) referenced in the body, or remove the references so navigation is not broken.
Add at least one concrete example — a sample tool call and a Python snippet — to satisfy the 'COMPUTE, DON'T DESCRIBE' principle and lift actionability.
Trim the redundant 'Summary' section and consolidate 'KEY PRINCIPLES' with the standalone directive sections to reduce token cost.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient with concrete phase/tool listings, but the terminal 'Summary' section restates the phases and the 'KEY PRINCIPLES' list overlaps with standalone sections like 'LOOK UP, DON'T GUESS' and 'COMPUTE, DON'T DESCRIBE'. | 3 / 5 |
Actionability | Concrete tool function names and a parameter table make the guidance largely executable, but no example tool calls or Python snippets are provided despite the skill's own 'COMPUTE, DON'T DESCRIBE' directive. | 4 / 5 |
Workflow Clarity | Phases 0–8 are clearly sequenced with 'ALWAYS FIRST' ordering, a report-first progressive-population loop, and a terminal completeness checklist, though inter-phase validation checkpoints are only implicit. | 4 / 5 |
Progressive Disclosure | References are clearly signaled and one level deep (phase-procedures.md, report-template.md, tool-reference.md, reference-data.md), but none of these bundle files actually exist in the skill directory, so the promised navigation is broken. | 3 / 5 |
Total | 14 / 20 Passed |