Conciseness

The body is mostly efficient, dense skill-specific guidance, but rules are repeated: the back-linking mandate appears in the Contract, the filing-rule quote, the Iron Law paragraph, and Output Format; "don't overwrite user-written assessments" appears in the UPDATE path, Validation Rules, and Anti-Patterns. The Philosophy section is motivational color rather than instruction. This fits 'mostly efficient but includes some unnecessary explanation or could be tightened'; it is above 2 because there is no padding explaining concepts Claude already knows.

Actionability

Copy-paste-ready person and company page templates, concrete commands ("gbrain search 'name'", "gbrain sweep --once"), a full JSON raw-data payload example, and explicit source tables cover the common cases. This matches 'fully executable; copy-paste ready code or commands; specific examples cover the common cases'; it is not 4 because the templates alone are directly usable and the commands are concrete, not pseudocode.

Workflow Clarity

A clearly sequenced 7-step protocol with CREATE/UPDATE branching, tier-scaled effort, validation rules (connection count, name mismatch), test-first bulk processing ("Test on 3-5 entities first"), and error-recovery feedback loops for rate_limited, revision_conflict, and auto_links.errors. This matches the top anchor despite batch operations, because validation checkpoints and fix-retry loops are explicitly present.

Progressive Disclosure

The single file is well-sectioned with clear headers, and references (skills/_brain-filing-rules.md, skills/conventions/quality.md, AGENT_OPERATOR_v1.md) are clearly signaled as one-level-deep pointers — though none are bundled with the skill. It fits 'good structure; most content is appropriately placed; minor organization gaps' rather than 5 because ~85 lines of page templates are inlined and could be split into reference files.

Specificity

"Creates and updates person/company pages with compiled truth, timeline, and cross-links" lists several concrete actions (create, update, compile truth, timeline, cross-links), matching the 'several specific actions; minor gaps' anchor. It stays at 4 rather than 5 because external research, tiering details, and raw-data provenance storage are part of the skill's work but go unmentioned.

Completeness

The description explicitly answers both questions: what ("Enrich brain pages... Creates and updates person/company pages with compiled truth, timeline, and cross-links") and when ("Use when a new entity is mentioned or an existing page needs updating") with concrete trigger phrasing, matching the top anchor. It is not 4 because the 'when' clause is already explicit and specific, not merely improvable.

Trigger Term Quality

Natural phrases like "enrich", "person/company pages", "a new entity is mentioned", and "existing page needs updating" are terms a user would plausibly say, fitting 'good keyword coverage; a few natural terms missing'. Not 5: common synonyms such as "profile", "contact", "look up", or specific mention of research are absent from the description itself.

Distinctiveness Conflict Risk

"Enrich brain pages" with "person/company pages" carves a fairly distinct niche with little overlap against unrelated skills, fitting 'mostly distinct; minor overlap risk'. It does not reach 5 because "creates and updates... pages" could still collide with generic page- or note-authoring skills absent the brain context.