Content
56%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 highly actionable with well-sequenced, validated workflows for schema reading and migrations, but it is bloated: repeated checklists, decorative ASCII diagrams, and full per-stack boilerplate that belongs in reference files. Restructuring into a lean core with per-stack references would substantially improve token efficiency and organization.
Suggestions
Move the per-stack type-generation code (Drizzle, Prisma, Supabase, SQLAlchemy) into separate reference files (e.g. references/drizzle.md, references/prisma.md) and keep only a stack-location table plus one-line pointers in SKILL.md.
Cut the decorative ASCII box diagrams (use plain numbered lists) and consolidate the three overlapping checklists (Pre-Code, session state, final Checklist) into one.
Trim boilerplate schema/model definitions Claude already knows — keep only project-specific conventions like the schema-reference.md template and the verification checklist format.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 380-line body includes decorative ASCII box diagrams, full boilerplate schema definitions for Drizzle/Prisma/SQLAlchemy that Claude already knows, and a verification checklist repeated three times (Pre-Code Checklist, Session Start state, final Checklist). This is noticeably verbose with several padded sections — anchor 2 — though not anchor 1 since there is no conceptual tutoring prose. | 2 / 5 |
Actionability | Concrete executable commands ('npx prisma generate', 'supabase gen types typescript --local > src/types/database.ts', 'npx drizzle-kit generate') and real type-safe query examples cover common cases, with minor gaps: the SQLAlchemy example uses 'relationship' without importing it, and the TypeORM row gets no example at all. Anchor 4 (mostly executable, minor gaps) rather than 5. | 4 / 5 |
Workflow Clarity | The migration workflow is a clear six-step sequence with explicit validation ('Run type check - find all broken code', 'Fix type errors, update tests, run full validation'), and the Schema Verification Checklist acts as a pre-code checkpoint — satisfying the database-operations feedback-loop requirement. Minor gaps ('If schema file doesn't exist → CREATE IT' with no how) keep it at anchor 4 rather than 5. | 4 / 5 |
Progressive Disclosure | No bundle files exist and everything is inline in one 380-line file; per-stack type-generation guides and the schema-reference template clearly belong in separate reference files. Section headers are clear and navigation within the file is fine, which lifts it above anchor 2 ('minimal structure') to anchor 3 ('some structure, content that should be separate is inline'). | 3 / 5 |
Total | 13 / 20 Passed |