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 has sound high-level sectioning (use/don't-use triggers, instructions, safety) but the instructions are abstract directives with no executable specifics, and the fractal reference structure is entirely broken — none of the 14 referenced sub-skill files exist in the bundle. Redundant Purpose/Philosophy sections and an empty Capabilities heading add tokens without adding information.
Suggestions
Make the Instructions actionable: add concrete decision criteria (e.g., a SQL vs NoSQL vs TimeSeries selection matrix), example schema/index/migration patterns, and specific validation commands instead of one-line directives like 'Choose the database model and architecture pattern'.
Fix progressive disclosure: either ship the 14 referenced sub-skills/*.md files in the bundle or inline their essential content into SKILL.md; annotate each link with a one-line note on when to consult it, and remove the empty '## Capabilities' heading.
Delete the redundant 'Purpose' section (it duplicates the frontmatter description almost verbatim) and fold the Safety items into the workflow as explicit validation checkpoints (e.g., 'validate migration plan in staging; only proceed to production rollout when validation passes').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 'Purpose' section restates the frontmatter description nearly verbatim ('Expert database architect with comprehensive knowledge of data modeling, technology selection...') and 'Core Philosophy' offers generic guidance Claude already knows ('Design the data layer right from the start to avoid costly rework'); the rest is lean but these padded sections and the empty '## Capabilities' heading could be trimmed. | 3 / 5 |
Actionability | The Instructions are high-level directives — 'Capture data domain, access patterns, and scale targets', 'Choose the database model and architecture pattern', 'Design schemas, indexes, and lifecycle policies' — with no concrete criteria, examples, commands, or decision methods for executing any step. This sits between anchor 2 (high-level hints, missing specific steps) and anchor 3, closer to 2 since no step includes executable specifics. | 2 / 5 |
Workflow Clarity | Four steps are listed in a sensible sequence and the Safety section mentions backups, rollbacks, and staging validation, but validation is not an explicit checkpoint within the workflow itself (no validate-then-proceed ordering). Per the rubric's cap for database operations involving destructive changes, workflow clarity cannot exceed 3. | 3 / 5 |
Progressive Disclosure | The body links to 14 sub-skill files (e.g. './sub-skills/technology-selection-evaluation.md', './sub-skills/data-modeling-schema-design.md'), but no sub-skills directory or any bundle files exist in the skill package — every reference is dead, the links carry no annotation about when to use each module, and '## Capabilities' is an empty heading. The reference structure does not resolve to any actual content, leaving navigation broken and the main body with almost no usable detail. | 2 / 5 |
Total | 10 / 20 Passed |