query

Answer questions using the brain's knowledge with 3-layer search, synthesis, and citation propagation. Use when the user asks a question, wants a lookup, or needs information from the brain.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

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 well-structured, actionable body with concrete gbrain commands, a clear phased workflow, and a useful error-recovery section. It is held back by noticeable duplication of source-precedence, hallucination, and citation guidance, and by the absence of any file-level progressive disclosure for its longer detailed sections.

Suggestions

Consolidate duplicated guidance: state source precedence, the no-hallucination rule, and citation format each in one canonical location instead of restating them across Contract, Anti-Patterns, Quality Rules, Output Format, and Citation in Answers.

Move the detailed Graph Traversal reference (link types, flags, and five worked examples) and the Search Quality Awareness troubleshooting into separate reference files (e.g. GRAPH.md, SEARCH-QUALITY.md) linked from the body, keeping SKILL.md a lean overview.

Integrate the search-quality check as an explicit validation checkpoint inside the Phases sequence rather than only as a separate troubleshooting section.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and avoids patronizing concept explanations, but the same guidance is restated multiple times — source precedence appears in Contract and again as a dedicated section, 'never hallucinate' in Contract, Anti-Patterns, and Quality Rules, and citation guidance across Output Format, Quality Rules, and Citation in Answers — so it could be tightened.	2 / 3
Actionability	It supplies concrete, copy-pasteable commands with real examples, e.g. 'gbrain graph-query companies/acme --type works_at --direction in' and 'gbrain doctor --json', matching the anchor for fully executable commands with specific examples.	3 / 3
Workflow Clarity	The Phases section gives a clear 5-step sequence (Decompose → Execute → Read top results → Synthesize → Flag gaps), with the Search Quality Awareness section providing an error-recovery feedback loop and step 5 acting as a verification checkpoint; this is a non-destructive lookup skill so the destructive-operation cap does not apply.	3 / 3
Progressive Disclosure	The body is well-organized into clear sections, but it is a single ~125-line file with no external references or file-level disclosure; the >50-line exemption for simple skills does not apply, and detailed sections like Graph Traversal and Search Quality Awareness are inline rather than split into reference files.	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 strong, third-person description that pairs concrete capabilities with an explicit Use-when clause and a clearly scoped niche. Its only weakness is that the vivid natural trigger terms ('tell me about', 'who is', 'what happened') are isolated in the frontmatter triggers list rather than woven into the description prose.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — '3-layer search, synthesis, and citation propagation' — matching the anchor for naming several specific concrete actions rather than vague language.	3 / 3
Completeness	It states what ('Answer questions using the brain's knowledge with 3-layer search, synthesis, and citation propagation') and when via an explicit 'Use when the user asks a question, wants a lookup, or needs information from the brain' clause.	3 / 3
Trigger Term Quality	The description's natural phrases ('asks a question', 'wants a lookup', 'needs information from the brain') are relevant but somewhat generic and lack common variations; the richer trigger terms live in the separate frontmatter triggers field rather than the description itself.	2 / 3
Distinctiveness Conflict Risk	References to 'the brain's knowledge', 'citation propagation', and niche triggers like 'graph query' and 'who knows who' carve a clear, distinctive niche unlikely to conflict with generic skills.	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/query/SKILL.md
Commit: 3fafb69

Reviewed: about 19 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.