enrich

Enrich brain pages with tiered enrichment protocol. Creates and updates person/company pages with compiled truth, timeline, and cross-links. Use when a new entity is mentioned or an existing page needs updating.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced enrichment protocol with concrete commands, templates, and validation gates for batch work. The main weaknesses are conciseness (redundant sections) and progressive disclosure (everything crammed into one large file rather than split into bundled references).

Suggestions

Remove the 'Output Format' section — it restates the person/company templates above it; the templates are the source of truth.

Tighten or cut the 'Philosophy' editorial paragraph, which adds no executable guidance.

Move the page templates and the data-source/tier tables into a bundled reference file (e.g. references/templates.md) and link from SKILL.md to reduce the monolithic inline content.

Dimension	Reasoning	Score
Conciseness	The ~320-line body is mostly substantive but carries redundancy: the 'Output Format' section restates the person/company templates above it, the 'Philosophy' paragraph is editorial padding, and 'Anti-Patterns' overlaps 'Do NOT enrich' and 'Validation Rules'. It is not a 3 because several passages could be tightened or cut, and not a 1 because it avoids explaining basic concepts Claude already knows.	2 / 3
Actionability	Concrete executable guidance throughout — `gbrain search`/`query` commands, a `put_raw_data` JSON example, full copy-paste person/company page templates, and named source APIs (Crustdata, Proxycurl, Perplexity) mapped to tiers — matching the fully-executable anchor.	3 / 3
Workflow Clarity	A clearly sequenced 7-step protocol with split CREATE/UPDATE paths, explicit validation (notability gate, connection-count<20, name-mismatch checks) and a batch checkpoint ('Test on 3-5 entities first... only proceed after test shots pass'). The validation gates and bulk checklist satisfy the feedback-loop requirement for batch/destructive work, so it is not capped at 2.	3 / 3
Progressive Disclosure	No bundle files exist and the skill is a single monolithic SKILL.md; external references (`skills/_brain-filing-rules.md`, `skills/conventions/quality.md`) are clearly signaled via blockquotes, but templates, tier/source tables, and rules that could live in separate reference files are all inline. It is not a 3 because content that should be split remains inline, and not a 1 because the inline references are one level deep and well signaled.	2 / 3
	Total	10 / 12 Passed

Description

85%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A clear, specific description that answers what and when with an explicit 'Use when' clause and a distinct niche. Its only weakness is trigger-term coverage: the natural user phrasings are isolated in the `triggers` field rather than woven into the description text.

Suggestions

Fold the natural trigger phrases from the `triggers` field (e.g. 'who is this person', 'look up this company') into the description so users' likely phrasings appear in the description itself.

Add common variations such as 'add to brain', 'enrich this contact', or 'company profile' to broaden natural keyword coverage within the description.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and deliverables — 'Creates and updates person/company pages with compiled truth, timeline, and cross-links' — matching the 'lists multiple specific concrete actions' anchor rather than the partial coverage of a 2.	3 / 3
Completeness	Explicitly answers both what ('Creates and updates person/company pages...') and when ('Use when a new entity is mentioned or an existing page needs updating'), satisfying the explicit-trigger anchor; the 'Use when...' clause prevents capping at 2.	3 / 3
Trigger Term Quality	The description itself surfaces only 'enrich', 'person/company pages', and 'entity'; the genuinely natural user phrasings ('who is this person', 'look up this company') live in the separate `triggers` field, so coverage is relevant but missing common variations. It is not a 3 because the natural keyword set is not present in the description text, and not a 1 because several relevant terms do appear.	2 / 3
Distinctiveness Conflict Risk	The brain/person-company enrichment niche with 'compiled truth, timeline, and cross-links' is specific and unlikely to fire for unrelated skills, matching the clear-niche anchor rather than the overlap-prone 2.	3 / 3
	Total	11 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: garrytan/gbrain
Path: skills/enrich/SKILL.md
Commit: 3fafb69

Reviewed: about 21 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.