Content
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is dense but purposeful: operational contracts, executable install/connection code, and a clear reference map with all referenced bundle files present. It avoids explaining concepts Claude already knows and structures detail into one-level-deep reference files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and operational — it does not explain what OMERO, BlitzGateway, or libraries are, and every section (Operating Contract, Install, Credentials, Helpers, Capability Guide) carries actionable load. It is tighter than the level-2 anchor that pads with "good library" justifications, and assumes Claude's competence throughout. | 3 / 3 |
Actionability | It provides copy-paste-ready, executable guidance: the `uv venv`/`uv pip install` install block, a complete exception-safe BlitzGateway Python snippet with try/finally, and concrete `python -B scripts/*.py --help` commands. This matches the level-3 anchor of fully executable, specific examples rather than pseudocode. | 3 / 3 |
Workflow Clarity | Risky remote and write workflows are clearly sequenced with explicit checkpoints: the eight-step Operating Contract, dry-run-by-default/`--execute` gating, the numbered Final Review checklist, and "Do not connect until the user has selected..." validation before remote work. This exceeds the level-2 anchor where validation is implicit. | 3 / 3 |
Progressive Disclosure | The SKILL.md is an overview with well-signaled one-level-deep references — the Capability Guide maps each topic to a real file in references/ (connection, data_access, metadata, image_processing, rois, tables, scripts, advanced, sources), and helpers live in scripts/, all verified to exist. This matches the level-3 anchor of an overview pointing cleanly to split reference files. | 3 / 3 |
Total | 12 / 12 Passed |