Content
100%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 content is an exemplar skill body: lean imperative instructions with zero padding, fully executable commands and templates, a clearly sequenced workflow with validation and feedback loops, and a clean overview-plus-references structure. Every dimension lands on the top anchor without ambiguity.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense, imperative, and assumes competence throughout — e.g. "Read product code only where a test needs a selector, a route, or a fixture" and "Never use `pkill -f` with a pattern that also appears in your own command line" — with no explanation of concepts Claude already knows. The few rationale sentences ("dev servers often listen on IPv6 only", "means nothing once the ticket closes") are short, non-obvious, and improve adherence rather than pad. | 5 / 5 |
Actionability | Fully executable guidance: copy-paste-ready commands (`pnpm exec playwright test <file> --timeout=15000 --max-failures=1`, `pytest tests/api/test_orders.py -x -k create`, `cmd & pid=$!`), complete markdown templates for test-plan.md and ledger.md, concrete naming examples (`live-updates.e2e.ts`, `screenshots/ac2-filter-after-reload.png`), and named harnesses (Django's test client, FastAPI's `TestClient`, `supertest`, Rails request specs) covering the common cases. | 5 / 5 |
Workflow Clarity | A clear sequence (read `.factory/CONTEXT.md` → sort criteria observable/internal → choose the cheapest layer → write one test per criterion → record plan and ledger → output the template) with explicit validation checkpoints ("Run only that test", "stop at the first failure") and feedback loops ("Fix the test with a small edit and run it again"; "A failure that shows the product misses the criterion is a finding, not something to fix"). Validation is present throughout, so the destructive/batch cap does not apply. | 5 / 5 |
Progressive Disclosure | The body is a well-organized overview that delegates detail through three clearly signaled, one-level-deep references — [references/api.md], [references/playwright.md], and [references/output.md] — each verified to exist and appropriately scoped (34, 29, and 26 lines). Content is split at the right boundaries: per-layer testing detail, tool-specific practices, and the output template all live outside the main file. | 5 / 5 |
Total | 20 / 20 Passed |