academic-verify

Verify a research claim or academic citation by tracing it through publication → methodology → raw data → independent replication. Routes through perplexity-research for the actual web lookup, then formats results as a citation-checked brain page. Use when a book/article/conversation cites a study and you want to confirm the claim is real, replicated, and accurately characterized.

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

75%Weight 40%Scale 1-5

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

A well-structured, highly actionable body with a concrete workflow and honest anti-pattern guidance. Its main weakness is redundancy — the verdict taxonomy is defined three times and two sections exist solely to satisfy a conformance test — plus a missing error-recovery loop in the workflow.

Suggestions

State the five-verdict taxonomy once (the 'Standards (the rigor bar)' section is the natural home) and reference it from Step 4 and the output template instead of repeating it three times.

Drop or merge the 'Contract' and 'Output Format' meta-sections that exist only for the conformance test — they add ~15 lines of non-operational tokens; fold any binding guarantees into the relevant workflow steps.

Add an explicit feedback loop to the workflow for the common failure case: when perplexity-research returns no original publication or no independent replication, specify whether to retry with adjusted search terms, record the verdict as 'Unverifiable', or both.

Consider moving the database URL table and the full output page template to a references/ file, keeping SKILL.md as the overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient but with real tightening opportunities: the five-verdict taxonomy is stated three times (Step 4, the 'Standards (the rigor bar)' section, and the output template), and the 'Contract'/'Output Format' sections explicitly exist 'for the conformance test' rather than for the operator. This is the 'could be tightened' anchor rather than 'minor instances that could be trimmed'.	3 / 5
Actionability	Fully executable for an instruction-only skill: a copy-paste gbrain query ('gbrain query "<paper title> OR <author name> OR <claim keywords>"'), a five-item checklist for the perplexity-research prompt, a complete markdown template for the output page, and concrete URL patterns for each database (e.g. 'retractionwatch.com/?s=NAME').	5 / 5
Workflow Clarity	A clear five-step sequence (scope → brain-first lookup → invoke perplexity-research → format verdict → cross-link) with an early reuse checkpoint in Step 2, and anti-pattern guards covering common failure modes. It stops short of the top anchor: no explicit feedback loop for what to do when the lookup returns nothing or contradicts an existing brain page.	4 / 5
Progressive Disclosure	Good structure with clearly-signaled one-level-deep links to conventions (quality.md, brain-first.md) and sibling skills at the top; no nested reference chains. Minor gaps: all reference targets live outside this bundle (unverifiable here), and the database table / output template are inlined where a references file could slim the main body.	4 / 5
	Total	16 / 20 Passed

Description

87%Weight 40%Scale 1-5

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 description: concrete actions, an explicit 'Use when' clause with natural trigger phrasing, and a well-delineated niche. Its only deductions are the second-person phrasing in the trigger clause and a few missing natural synonyms.

Dimension	Reasoning	Score
Specificity	Multiple concrete actions are named — 'tracing it through publication → methodology → raw data → independent replication', 'Routes through perplexity-research for the actual web lookup, then formats results as a citation-checked brain page' — giving comprehensive coverage. A point is deducted per the guideline penalizing second-person voice ('you want to confirm the claim is real'), which the description uses in its trigger clause.	4 / 5
Completeness	Explicitly answers both: what ('Verify a research claim or academic citation by tracing it through publication → methodology → raw data → independent replication... formats results as a citation-checked brain page') and when ('Use when a book/article/conversation cites a study and you want to confirm the claim is real, replicated, and accurately characterized') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good natural keyword coverage: 'verify', 'research claim', 'academic citation', 'cites a study', 'real', 'replicated'. A few natural terms users might say are missing ('paper', 'journal', 'is this citation accurate', 'fact-check this study'), so it falls just below the comprehensive-with-synonyms anchor.	4 / 5
Distinctiveness Conflict Risk	Clear niche — verifying cited academic claims — with triggers that would not fire for generic document, code, or research skills; it even scopes the web-lookup portion to the separate perplexity-research skill, minimizing overlap with open-ended research. Minor residual overlap with a fact-checking skill is negligible.	5 / 5
	Total	18 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing, 3 suspicious	Warning

	Total	14 / 16 Passed

Repository: garrytan/gbrain
Path: skills/academic-verify/SKILL.md
Commit: e78f1c3

Reviewed: 5 days 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.