Conciseness

The body is lean with no concept-explaining padding, but content is duplicated: the "Mindset" two-question decision tree restates the "Tier selection" table ("Is this time-bound? If yes → episodic" vs "tied to today's session or a specific event? Yes → episodic"), and the quoting rule appears twice ("Always double-quote the --text value to handle apostrophes" and later "Never use single quotes around --text — apostrophes in content will break the shell command"). These are minor trimmable instances, so anchor 4 rather than 5's 'every token earns its place'; not anchor 3 since there is no unnecessary explanation of things Claude already knows.

Actionability

Quotes: the complete command signature "vault-cli capture --text \"<memory content>\" [--tier episodic|semantic|procedural] [--project <id>] [--tags tag1,tag2]", the stdin variant "echo \"<memory content>\" | vault-cli capture", and three fully-populated example commands covering semantic, procedural, and episodic cases. Copy-paste ready with specific examples covering the common cases — anchor 5, not anchor 4's 'minor gaps'.

Workflow Clarity

This is a simple, single-action skill: a two-question decision procedure ("Is this time-bound? If yes → episodic ... Is this a how-to process? If yes → procedural. If no → semantic.") selects the tier, then one unambiguous capture command runs with the required flags. Capture is a single non-destructive append, so the destructive/batch validation cap does not apply, and per the simple-skill note the unambiguous single action earns 5; the sequence is coherent and free of gaps that would drop it to 4.

Progressive Disclosure

The body is ~62 lines, self-contained, with no bundle files and no external references — nothing is inlined that belongs in a separate file, and there are no nested/buried references. Sections (Mindset, When to use, How to use, Tier selection, Examples, Never) are well-organized and easy to navigate, which per the simple-skill scoring note merits 5 for a self-contained skill of this size; not 4 because there are no organization gaps to point to.

Specificity

Quotes: "Captures an important insight, decision, constraint, pattern, or discovery to persistent memory via the vault CLI" and "entries are tagged, tiered, and retrievable across future sessions". It names the tool, the concrete action, and the mechanism (tagged/tiered/retrievable) with enumerated capture types. It stops short of anchor 5 because it presents one core action with rich modifiers rather than several distinct actions, and the tier taxonomy is only hinted at ("tiered"), leaving minor coverage gaps — better than anchor 3's '1-2 concrete actions, not comprehensive'.

Completeness

The 'what' is explicit ("Captures ... to persistent memory via the vault CLI", "tagged, tiered, and retrievable") and the 'when' is explicit with concrete trigger phrases ("Use when the user says ... or when the conversation produces something worth preserving: ..."). This matches the anchor-5 example pattern exactly; it is not anchor 4 because both what and when are fully explicit and specific rather than 'could be more explicit'.

Trigger Term Quality

Quotes: "save this", "remember this", "note for later", "don't forget", "keep this in mind", plus content-category triggers "architectural decisions, recurring bugs, hard-won constraints, naming conventions, or user preferences". Five distinct natural user phrasings plus synonyms cover the domain comprehensively (no file extensions apply to a memory-capture skill), matching anchor 5 rather than anchor 4's 'a few natural terms missing'.

Distinctiveness Conflict Risk

Quote: "distinct from general note-taking in that entries are tagged, tiered, and retrievable across future sessions" — it explicitly differentiates from the nearest competing skill class, and the vault CLI + tier/tag system carves a clear niche. Minimal conflict risk; better than anchor 4's 'minor overlap risk with closely related skills'.