Conciseness

Lean and efficient: every section carries non-obvious, Prism-specific knowledge ('--multiprocess=false avoids a createMultiProcessPrism crash', 'with default --errors=false Prism validates only the spec's required fields') rather than concepts Claude already knows. Not 4: there is no trimming opportunity that wouldn't lose operative detail.

Actionability

Mostly executable: a copy-paste-ready 'prism mock <spec> --port 13000 --cors --multiprocess=false' command, a kill/lsof cleanup pair, and a harness usage snippet whose API (rewriteUrl/auth/watch/transformBody, h.route/h.restore) matches the actual scripts/prism-harness.mjs. Not 5: the inline JS snippet has minor gaps — 'base' is referenced inside its own object literal (would throw if run verbatim) and 'token' is undefined — and full runnable code is deferred to the reference file.

Workflow Clarity

Four clearly sequenced steps, each ending in an explicit 'Done when:' validation checkpoint, with error-recovery feedback ('A crash means add --multiprocess=false; a taken port means pick another'), a red/green loop in step 3 ('one command goes red on the bug and green once fixed'), and a post-run port verification via 'lsof -i :<port>'. Not 4: checkpoints and recovery paths are present at every stage.

Progressive Disclosure

Clear overview with a well-signaled, one-level-deep reference — [references/coolisimplify-example.md] with explicit when-to-read guidance ('Read it when the task touches this repo's clone/API flows') — plus a real, matching scripts/prism-harness.mjs (both files verified to exist, and the harness API matches the script). Content is appropriately split for a skill of this size. Not 4: no organization gaps; navigation is easy and nothing that belongs in a separate file is inlined.

Specificity

Lists multiple concrete actions — 'Test a client against a Prism mock server built from an OpenAPI spec', 'drive a client or harness through an OpenAPI spec', 'verify request payloads or 422 behavior' — comprehensively covering the skill's scope. Not 4: there is no meaningful coverage gap; the actions are specific and complete for this domain.

Completeness

Explicitly answers both: what — 'Test a client against a Prism mock server built from an OpenAPI spec'; when — 'Use when the user wants to test against a mock API, drive a client or harness through an OpenAPI spec, verify request payloads or 422 behavior against a spec, or when a task needs a local fake of a REST API for fast, deterministic testing'. Not 4: the 'when' clause is highly explicit with multiple concrete trigger phrases, not merely adequate.

Trigger Term Quality

Good natural-term coverage including synonyms: 'mock API', 'local fake of a REST API', 'OpenAPI spec', 'request payloads', '422 behavior', 'fast, deterministic testing'. Not 5: a few natural terms users might say are missing (e.g. 'stub', 'fake server', 'Swagger').

Distinctiveness Conflict Risk

Clear niche (Prism mock server generated from an OpenAPI spec) with distinct, specific triggers ('Prism mock server', 'OpenAPI spec', '422 behavior') that are unlikely to fire for the wrong skill. Not 4: overlap with generic API-testing or mocking skills is minimal given the spec-driven framing.