Content
57%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-organized, code-heavy Django patterns reference with strong, mostly executable examples, but it is monolithic (~730 lines, no external reference files), repeats guidance across sections, and includes batch operations with no validation steps. Splitting detail into reference files and adding verification guidance for batch/destructive operations would raise the score.
Suggestions
Split detail into one-level-deep reference files (e.g. references/settings.md, references/drf.md, references/caching-signals.md, references/performance.md) and keep SKILL.md as a concise overview with clearly signaled links, per progressive disclosure.
Add validation/verification guidance to the batch operations section (e.g. verifying counts or wrapping bulk operations in transaction.atomic with checks), since batch and destructive operations currently lack any feedback loop and cap workflow clarity at 3.
Trim duplicated content — the N+1 section restates select_related/prefetch_related already demonstrated in the QuerySet and ViewSet sections, and the database index section repeats Meta.indexes from the Product model — and fix non-executable snippets (undefined env(), slugify, logger, timezone, Count, PaymentGateway, Cart).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is code-dense with almost no prose fluff, but at ~730 lines it largely restates standard Django patterns (custom user model, QuerySet, ViewSet, signals, middleware) Claude already knows, and duplicates content — the N+1 section repeats select_related/prefetch_related guidance already shown in the QuerySet and ViewSet sections, and the index section repeats Meta.indexes from the Product model. It is not 2 because there is little padded explanation, and not 4 because the duplication and re-teaching of well-known patterns are more than minor. | 3 / 5 |
Actionability | Most code blocks are near copy-paste-ready, production-quality Django (split settings, custom QuerySet/Manager, serializers with validation, ViewSet with filtering, service layer with transaction.atomic). It is not 5 because several snippets will not run as written: base.py uses an undefined env() with no django-environ import/setup, save() uses slugify without import, the middleware references undefined logger/timezone, and the service layer references undefined PaymentGateway, Cart, and Count — alongside '# ... fields ...' ellipses. Not 3, since the gaps are import-level omissions rather than missing steps or pseudocode. | 4 / 5 |
Workflow Clarity | The content is organized into clear topical sections (project structure, models, DRF, services, caching, signals, middleware, performance), but it is a pattern catalog with no sequenced workflow and no validation or verification steps anywhere — notably the 批量操作 section performs bulk create/update/delete with no checks, which caps workflow clarity at 3 per the batch-operations rule. Not 4 because validation checkpoints are entirely absent, not just minor. | 3 / 5 |
Progressive Disclosure | Section headers (##) provide genuine structure and navigation within the file, but this is a single monolithic ~730-line SKILL.md with no references/ or scripts/ bundle — detailed settings, DRF, caching, signals, and performance content clearly belongs in separate one-level-deep reference files. It is not 2 because the section headers give real structure (unlike the header-less inlined blob anchor), and not 4 because nothing is split out into external files. | 3 / 5 |
Total | 13 / 20 Passed |