Content
86%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.
A well-structured, highly actionable skill body with excellent progressive disclosure via a phase-to-reference table and copy-paste-ready commands. Minor gains remain from deduplicating repeated constraints and making the validation feedback loop explicit.
Suggestions
Consolidate the app-name rules (stated in Generic Guidelines, Scaffolding 'Required', and Directory Naming) into one canonical location and cross-reference it.
Add an explicit validation feedback loop after `databricks apps validate` (e.g., 'If validation fails, fix the reported errors and re-run validate before deploying').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean directives, tables, and executable commands that assume Claude's competence; not a 5 because the app-name constraint and 'DO NOT guess' messaging are repeated 2-3 times across sections and could be consolidated. | 4 / 5 |
Actionability | Provides copy-paste-ready commands (`databricks apps init --name ... --features ... --set ...`, `databricks apps manifest`, `npx @databricks/appkit docs`, `npm run typegen`, `databricks apps validate`) with right/wrong examples covering common scaffolding cases. | 5 / 5 |
Workflow Clarity | The 'Development Workflow (FOLLOW THIS ORDER)' gives a clear numbered sequence with a typegen verification checkpoint and a final `databricks apps validate` step; not a 5 because an explicit validate-fix-retry feedback loop is not spelled out. | 4 / 5 |
Progressive Disclosure | A 'Required Reading by Phase' table maps each phase to a specific one-level-deep reference file, all of which exist, and content is appropriately split across references/appkit/* and references/* with easy navigation. | 5 / 5 |
Total | 18 / 20 Passed |