Content
46%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 well-organized and token-efficient with sensible use/don't-use boundaries, but it fails at its main job as an overview hub: the only reference points to a nonexistent resources/implementation-playbook.md while all four actual bundle files go unmentioned, and the Instructions themselves are too abstract to act on without those missing details. Fixing the reference paths and listing the real bundle files would resolve the biggest weaknesses.
Suggestions
Replace the broken `resources/implementation-playbook.md` reference with the actual bundle paths and enumerate all four files in the Resources section (e.g., "references/ef-core-best-practices.md" for EF Core guidance, "references/dapper-patterns.md" for Dapper guidance, "assets/repository-template.cs" and "assets/service-template.cs" as starting templates) so each is discoverable from SKILL.md.
Make the Instructions concrete by mapping each bullet to the material that implements it, e.g., "For data access, start from assets/repository-template.cs and pick EF Core vs Dapper using the comparison table in references/dapper-patterns.md" instead of the abstract "Validate data access performance and caching."
Give the Instructions an explicit sequence with checkpoints, e.g., order the bullets as a lifecycle (architecture → DI/async → data access → tests) and state when to verify, such as "validate query performance before introducing caching".
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean (~33 lines), assumes Claude's competence, and explains nothing the model already knows; every section carries information. Not anchor 5 because of minor redundancy: the playbook path is stated twice (Instructions bullet and Resources section) and the "Use this skill when" list partially restates the frontmatter description's scope. | 4 / 5 |
Actionability | The Instructions are abstract imperatives — "Define architecture boundaries, modules, and layering", "Apply DI, async patterns, and resilience strategies", "Validate data access performance and caching" — with no concrete steps, commands, patterns, or file-level pointers, matching anchor 2's "high-level hints but missing the specific steps to execute". Not anchor 3 because the body contains no concrete executable guidance at all; all specifics are deferred to a reference file, and not anchor 1 because the bullets do direct Claude's actions rather than merely describing. | 2 / 5 |
Workflow Clarity | The Instructions bullets form an implicit lifecycle sequence (architecture boundaries → DI/async/resilience → data access and caching validation → tests/observability) with a conditional escalation ("If detailed patterns are required, open..."), fitting anchor 3's "steps listed but validation gaps; checkpoints missing or implicit". Not anchor 4 because no explicit ordering or verification checkpoints are stated, and not anchor 2 because the steps themselves are clearly enumerated and "Validate data access performance" gestures at a checkpoint. | 3 / 5 |
Progressive Disclosure | Scored against the actual bundle: the body's sole reference, "resources/implementation-playbook.md", does not exist (no resources/ directory or playbook file is present), while the four real bundle files (references/dapper-patterns.md, references/ef-core-best-practices.md, assets/repository-template.cs, assets/service-template.cs) are never mentioned, leaving them undiscoverable — the "references are buried" failure of anchor 2. Not anchor 1 because the body itself is well-sectioned and short rather than a monolithic wall of text, and not anchor 3 because the one reference that is signaled is broken and the entire real bundle is orphaned, so navigation to any detailed content fails. | 2 / 5 |
Total | 11 / 20 Passed |