Conciseness

The body is lean and imperative throughout ('Add no finding the tool did not emit', 'Quote the entry body verbatim', 'Report the exit code and the error verbatim') with zero concept padding. The long reviewer-checklist example is a concrete output template, not filler, so every token earns its place — matching the 5 anchor rather than the 4 anchor's 'minor instances of over-explanation'.

Actionability

Fully executable guidance: exact commands with flags (`spectrace corpus review <spec-path> --format json`), exit-code semantics, two verbatim error strings with their meanings, exact output table column headers, a `grep -rl` lookup for corpus files, a worked checklist example, and a per-finding fix table. This meets the 5 anchor's 'copy-paste ready commands covering the common cases'.

Workflow Clarity

The four-step workflow is clearly sequenced with JSON as the declared source of truth, and step 4 is an explicit validation checkpoint ('Every finding in the JSON appears in your output. Nothing else does.'). Error recovery is explicit ('print the error verbatim and stop') with the UnknownCitationError-vs-orphan_citation distinction spelled out. This is a read-only review skill, so the destructive/batch validation cap does not apply; the sequence matches the 5 anchor's explicit checkpoints and error-feedback handling.

Progressive Disclosure

Sections are well-organized and the Reference section clearly signals one-level-deep docs with one-line descriptions, matching the 4 anchor's 'good structure; references mostly clear; minor organization gaps'. It falls short of the 5 anchor because the skill is a single 145-line inline body (the output-format spec and fix table could live in a reference file), it exceeds the under-50-line simple-skill exemption, and the referenced `docs/corpus-review.md` / `docs/corpus-authoring.md` are repo-relative paths not present in the skill bundle, so their content cannot be verified.

Specificity

Names multiple concrete actions with the exact tool: 'Review a spec in this repo against the SpecTrace corpus of org standards, decisions, and commitments', 'Runs `spectrace corpus review` and formats what it emits', and 'find unaddressed obligations or stale citations'. Coverage is comprehensive with no gaps that would pull it to the 4 anchor.

Completeness

Explicitly answers both questions: the 'what' is concrete (run `spectrace corpus review` and format its output against the org corpus), and the 'when' is an explicit 'Use when' clause with concrete trigger phrases and verbatim example utterances, mirroring the 5-anchor example pattern. Voice is third person ('Review', 'Runs'), so no voice penalty applies.

Trigger Term Quality

Includes quoted natural phrases a user would actually say: "corpus review this", "which standards apply to this spec", "is this spec covered", plus task-language synonyms ('review a spec against the corpus', 'check which standards a spec touches'). This is comprehensive coverage, not the 4 anchor's 'a few natural terms missing'.

Distinctiveness Conflict Risk

Clear niche (spec review against a versioned corpus of org standards/decisions/commitments) with a distinct tool name and niche-specific trigger phrases, so overlap with other skills is minimal and it would not fire for the wrong skill.