Conciseness

The body is lean and imperative throughout — "Treat the installed product as code and the selected user vault as data", "Fail closed when selection is missing or ambiguous", "Raw source payloads are create-only" — with zero background explanation of Obsidian, Git, or vaults that Claude already knows. It matches anchor 5 (assumes Claude's competence, every token earns its place) rather than anchor 4, which expects over-explanation worth trimming; none is evident.

Actionability

Concrete, executable commands cover the common cases: `python3 "$CORE" --help`, full init/adopt dry-run and `--apply` invocations with explicit flags, and `python3 "$CORE" checkpoint OPERATION_ID --vault ...`, plus an intent-to-skill routing table and a fail-closed vault resolution order. Placeholder values have explicit provenance ("pass the emitted `approved_plan_sha256` unchanged"), matching anchor 5; anchor 4 would require missing key details, and none are apparent.

Workflow Clarity

The setup flow has an explicit validation sequence (dry-run by default, "Before `--apply`, show the selected path and changed-path preview, then pass the emitted `approved_plan_sha256`"), destructive paths are gated ("Do not use `--force` unless the user has reviewed the conflicts and explicitly approved replacement"), and error recovery is explicit ("On a conflict, re-read and rebuild. On interruption, use `transaction recover`"). This is anchor 5's clear sequence with validation checkpoints and feedback loops, not anchor 4, which would have minor validation gaps.

Progressive Disclosure

The body is a well-signaled overview with one-level-deep references — "Read only the reference needed for the current request" with 14 conditionally-scoped links, all of which exist in references/ — which approaches anchor 5. However, three references point to `../../docs/native-conformance.md`, `../../docs/capacity-and-recovery.md`, and `../../docs/upstream-references.md`, which do not exist in the provided bundle structure, a minor navigation gap that anchor 4 ('minor organization gaps') captures better than anchor 5.

Specificity

"Initialize, adopt, and route work for a separate Obsidian knowledge vault" plus "vault setup, scaffolding, workspace selection, cross-project configuration, or choosing the correct wiki sub-skill" lists several specific, concrete capabilities for an orchestration skill. It falls short of anchor 5 because "route work" and the sub-skill list are abstract — the concrete actions the sub-skills perform are not stated — and it is clearly above anchor 3, which covers only 1-2 concrete actions.

Completeness

"Initialize, adopt, and route work ... through the portable claude-obsidian core" explicitly states what the skill does, and "Use for vault setup, scaffolding, workspace selection, cross-project configuration, or choosing the correct wiki sub-skill" plus an explicit trigger list states when to use it with concrete trigger phrases. This matches anchor 5 exactly (both what and when with concrete triggers); anchor 4 would require the 'when' to be less explicit.

Trigger Term Quality

"Triggers: /wiki, set up wiki, scaffold vault, create knowledge base, adopt this vault, Obsidian vault, second brain setup, persistent wiki" covers the natural phrasings users would say, including slash-command form and synonyms (vault, knowledge base, second brain, wiki). Better fits anchor 5's comprehensive synonym coverage than anchor 4's 'a few natural terms missing'; there are no file extensions in this domain to include.

Distinctiveness Conflict Risk

"separate Obsidian knowledge vault" and "portable claude-obsidian core" carve out a clear niche with mostly distinct triggers (Obsidian vault, adopt this vault, /wiki). It is not anchor 5 because generic triggers like "create knowledge base" and "persistent wiki" carry minor overlap risk with other knowledge-management or memory skills; it is well above anchor 3 since the domain and product are explicitly named.