Conciseness

The body is dense with project-specific knowledge Claude cannot know (the 401-sweep rationale, scratch naming rules, xfail policy) and explains no general concepts. Minor trimmable repetition exists — e.g. the xfail section restates "never soften the assertion to match the bug" twice in slightly different forms — placing it at the 4 anchor ("minor instances of over-explanation that could be trimmed") rather than 5.

Actionability

Fully executable throughout: a copy-paste module docstring template, an assert line with the diagnostic message format, the exact `@pytest.mark.requires_seed(...)` decorator, and exact commands (`./dev-compose.sh test-stand test --tree=...`, `uv run --project tests --frozen pytest/ruff`, `generate_schemas.py --check`, the coverage.py reconcile invocation) with flags explained. Matches the 5 anchor — copy-paste ready and covering the common cases.

Workflow Clarity

The 9-step Procedure is clearly sequenced with explicit validation checkpoints: run and check the ledger, ruff check, the schema drift-gate `--check`, and reconciling the coverage gate, with error-handling rationale ("a failing run's ledger is the more useful one"). Feedback loops (probe 422-vs-400 rather than assume, reconcile the gate after running) match the 5 anchor.

Progressive Disclosure

Well-organized single-file skill with clear section headers, tables, and clearly signaled cross-references to sibling skills (`insight-stand`, `stand-scenarios`, `metric-test`, `quality-vector-tests`, `file-bug-insight`) — no buried or nested references. However, the under-50-line simple-skill exception does not apply at ~280 lines, and content like the full fixture table and status-code matrix could live in a reference file, so the 4 anchor ("good structure; minor organization gaps") fits better than 5.

Specificity

Opens with three concrete verbs ("Write, fix, or review HTTP contract tests") and enumerates comprehensive coverage areas — operation catalogue, persona session choice, requires_seed markers, scratch-resource policy, hand-written vs generated models, status-code discipline (404/403/400/415/422), and the coverage gate. Multiple specific concrete actions with no coverage gaps; matches the 5 anchor rather than 4.

Completeness

Explicitly answers both questions: the "what" ("Write, fix, or review HTTP contract tests in tests/stand/api/ — the deployed-stand suite…") and a fully explicit "when" with concrete trigger phrases. Clearly the 5 anchor; the 4 anchor requires a 'when' that is only partly explicit, which is not the case here.

Trigger Term Quality

Natural trigger phrasings users would say are present: "Use when adding or changing anything under tests/stand/api/, closing a coverage-gate gap, or turning a stand-scenarios claim into an API case", plus concrete keywords ("contract tests", "coverage-gate gap", "Keycloak sessions", "endpoint coverage gate"). Coverage is comprehensive for this domain, matching the 5 anchor rather than 4 where only a few natural terms would be missing.

Distinctiveness Conflict Risk

Clear niche (the deployed-stand API suite) with active disambiguation from adjacent skills: "For browser journeys use stand-ui-test" and "Seeded-bronze → served-metric specs are the data-path suite in tests/datapath/ (metric-test)". Minimal conflict risk; matches the 5 anchor.