research-review

Get a deep critical review of research from an external reviewer backend (Codex or manual). Use when user says "review my research", "help me review", "get external review", or wants critical feedback on research ideas, papers, or experimental results.

Quality

82%

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

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.

The body is a well-structured, highly actionable external-review workflow with concrete tooling and clear sequencing. Its main weakness is conciseness — repeated Codex/manual branching and config blocks add tokens that could be consolidated — and lighter error-recovery feedback loops.

Suggestions

Consolidate the repeated Codex-vs-manual model/effort config and 'attach or paste the brief' guidance into one place (e.g. the Reviewer Calling Convention) and reference it, instead of restating in Workflow and Key Rules.

Add an explicit error-recovery / fallback loop for backend failures (e.g. Codex call fails → retry once → if still failing, switch per reviewer-routing.md → report) to strengthen workflow_clarity beyond a single 'stop and print install command' directive.

Move the inlined Review Tracing policy detail and composed-mode rules fully into their referenced shared-references files, leaving SKILL.md as a concise pointer, to tighten progressive_disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's intelligence (no basic-concept padding), but the Codex-vs-manual branching and model/effort config blocks are repeated across the Calling Convention, Workflow, and Key Rules sections, and the 'attach or paste the brief' guidance is stated multiple times — it could be tightened without losing signal.	3 / 5
Actionability	Provides exact MCP tool names, copy-pasteable config JSON (e.g. {"model_reasoning_effort": "ultra"}), concrete prompts, named brief files, a bash install command, and ready prompt templates covering the common review cases.	5 / 5
Workflow Clarity	A clear 5-step sequence (Gather, Initial Review, Iterative Dialogue, Convergence, Document) with explicit stop criteria and a tracing checkpoint; not a 5 because error-recovery feedback loops are light (e.g. backend unavailable just says stop and print the install command) and there is no explicit retry/repair loop.	4 / 5
Progressive Disclosure	Detail is deferred to clearly signaled one-level-deep shared-references (external-cadence.md, reviewer-routing.md, output-composition.md, review-tracing.md, integration-contract.md), keeping the overview in SKILL.md; not a 5 because some reference-targeted rules (tracing detail, composed-mode rules) are inlined and the referenced shared-references files are not present in this bundle to verify.	4 / 5
	Total	16 / 20 Passed

Description

90%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.

The description is strong: it clearly states the what and when with concrete, natural trigger phrases and a distinct niche. The only weaker dimension is specificity, where the single 'get a review' action keeps it from listing multiple concrete capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and mechanism ("deep critical review of research from an external reviewer backend (Codex or manual)") but the action set is essentially singular — get a review — without enumerating multiple distinct concrete actions, fitting the 'names domain and 1-2 concrete actions' anchor rather than the 'several specific actions' level.	3 / 5
Completeness	Explicitly answers both what ("Get a deep critical review of research from an external reviewer backend") and when ("Use when user says...") with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Multiple natural phrases users would actually say ("review my research", "help me review", "get external review") plus synonyms (critical feedback on research ideas, papers, experimental results) give comprehensive coverage of the domain's trigger surface.	5 / 5
Distinctiveness Conflict Risk	The external cross-model reviewer backend framing and research-specific triggers carve a clear niche with minimal overlap risk against other skills.	5 / 5
	Total	18 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/research-review/SKILL.md
Commit: f4f20f9

Reviewed: 1 day 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.