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 is long and boilerplate-heavy, re-explains statistical concepts Claude already knows, and ships broken cross-references plus references to non-existent bundle files. Concrete argument/usage tables provide some actionability, but corrupted parameter tables and commented-only examples leave guidance incomplete.
Suggestions
Remove the generic boilerplate sections (Risk Assessment, Security Checklist, Evaluation Criteria, Lifecycle Status, Output Contract) and the re-explanation of KM/log-rank/Cox basics to cut the body to a lean overview.
Fix the duplicate/corrupted Parameters table (empty descriptions, `--figsize` str '10', `--risk-table` Required vs Optional) and provide a real, runnable example command instead of commented `# Example invocation` lines.
Resolve the broken 'See `## X` above' pointers (Features/Usage/Workflow appear below, not above) and either ship the referenced `scripts/main.py` and `references/` or stop pointing to them.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | At ~320 lines the body is padded with generic boilerplate (Risk Assessment, Security Checklist, Evaluation Criteria, Lifecycle Status, Output Contract, etc.) and re-explains concepts Claude already knows (KM estimator, log-rank, Cox model), with several duplicated/broken 'See ## X above' pointers — matches 'noticeably verbose; several unnecessary explanations or padded sections'. | 2 / 5 |
Actionability | There are concrete argument tables and example invocations, but the runnable examples are only commented `# Example invocation` lines, the duplicate Parameters table is corrupted (empty descriptions, `--figsize` str '10', `--risk-table` listed Required), and the referenced `scripts/main.py` does not exist — concrete but incomplete. | 3 / 5 |
Workflow Clarity | Two competing workflow lists exist with a py_compile validation checkpoint and a fallback path, but steps are generic and duplicated, cross-references are broken, and validation is implicit rather than a tight validate→fix→retry loop. | 3 / 5 |
Progressive Disclosure | The body points to `scripts/main.py` and `references/` for papers, datasets, and guidelines, but neither bundle directory exists, and navigation is broken ('See `## Features` above' when Features is below) — references are non-functional and much generic content is inlined. | 2 / 5 |
Total | 10 / 20 Passed |