Conciseness

The body is dense and operational — preset mapping tables, exact commands, gates, error tables — with almost no explanation of concepts Claude already knows. Minor instances could be trimmed: the five near-identical Invocation Examples, the repeated consent-gate rules ('not to clear a refused recording, not as an error-recovery retry, not to save a round-trip'), and the Scope section restating the description. This matches 'Efficient; minor instances of over-explanation that could be trimmed' rather than level 5, where every token earns its place.

Actionability

Fully executable, copy-paste-ready commands throughout: 'command -v xcprof && xcprof doctor', 'export XCPROF_TRACE_ROOT="$(mktemp -d)"', simctl discovery with jq, a user-intent-to-invocation table with exact flags and time limits, and a symptom/cause/fix error table. Parameter placeholders ('<app>', "<trace>") are appropriately explicit. This matches 'Fully executable; copy-paste ready code or commands; specific examples cover the common cases'.

Workflow Clarity

A clear four-step sequence (pick target → record → analyze → report) with explicit validation checkpoints: 'doctor' readiness verification, '--dry-run' preview, reading the ok/notes fields, and the support-status honesty gate before calling anything clean. Feedback loops are present in the Error handling table (symptom → cause → fix) and the hang-window re-scope loop. Destructive operations are explicitly guarded (no 'rm -rf' of trace dirs; consent required before gated flags). This matches 'Clear sequence with explicit validation steps; feedback loops for error recovery'.

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent), and the body's pointers — 'axiom-tools (skills/xcprof-ref.md)', 'axiom-performance (skills/trace-comparison.md)', etc. — are clearly signaled one-level-deep cross-skill references collected in a Related section. Structure is good with well-labeled sections. It is not level 5 because some inlined content (the full Comparison/CI-gating section, the Invocation Examples block, detailed flag semantics) could be split out or trimmed, leaving minor organization gaps; it is well above level 3 since nothing is buried and navigation is easy.

Specificity

The description names its domain ("automated performance profiling, headless Instruments analysis, or CLI-based trace collection") but states no actions the skill performs (e.g., recording traces, analyzing them, producing reports) — noun phrases describing the domain, matching the anchor 'Names the domain but actions are minimal or generic'. It is not level 3 because a level-3 description would list 1-2 concrete actions the skill carries out, and here the 'what' is only inferred from the trigger clause.

Completeness

Only the 'when' is present ("Use when the user wants automated performance profiling...") without any explicit 'what' — the description never states what the skill does, exactly matching the anchor branch 'only when is present without what'. It is not level 3 because that anchor requires a clear 'what' (like 'Extract text and tables from PDF files'), which is only weakly inferable here; scoring only what is explicitly stated, the 'what' is missing.

Trigger Term Quality

Relevant keywords are present ("performance profiling", "Instruments", "trace") but common natural variations users actually say are missing — e.g. "CPU", "memory leaks", "Time Profiler", "slow app", "optimize" — matching 'Some relevant keywords but missing common variations or synonyms'. It is not level 4 because the coverage gaps are more than 'a few' (the words users most often lead with, like CPU or leaks, are absent); it is above level 2 because multiple domain-specific natural terms (Instruments, profiling, trace) do appear.

Distinctiveness Conflict Risk

The description occupies a clear niche — headless Apple Instruments/xctrace profiling via CLI — with distinct triggers and minimal overlap risk with generic profiling or analysis skills. It matches 'Clear niche with distinct triggers; minimal conflict risk' exactly.