research-review

Get a deep critical review of research from GPT using a secondary Codex agent. 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

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/skills-codex/research-review/SKILL.md

Quality

Content

67%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 content is a well-structured, actionable review workflow with clear sequencing and concrete agent-invocation templates. Its main weakness is conciseness, driven by duplicated prompt material between the workflow steps and the Prompt Templates section.

Suggestions

Remove the standalone Prompt Templates section or convert it to a pointer, since the same prompts already appear inline in Steps 2-3; this eliminates the largest source of redundancy.

Condense the Codex-assurance blockquote and same-family routing notes into a single short line, moving the full routing rationale into the already-referenced reviewer-routing.md.

Add an explicit validate/feedback step (e.g., "re-run the reviewer on the revised materials and confirm each prior weakness is resolved") to push workflow clarity toward 5.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and assumes Claude's competence, but the Prompt Templates section duplicates the prompts already embedded in Steps 2-3 and the Codex-assurance/routing notes are padded, so it could be tightened rather than being fully lean.	3 / 5
Actionability	Provides concrete spawn_agent/send_input blocks with model, reasoning_effort, and message fields plus specific follow-up patterns, but uses placeholders ([Full research context...], /absolute/path/to/file1) so it is not fully copy-paste ready.	4 / 5
Workflow Clarity	A clear six-step sequence with an explicit convergence checkpoint in Step 4 and documentation/tracing steps; this is not a destructive or batch operation so the validation cap does not apply, but there is no hard validate-fix-retry feedback loop to reach 5.	4 / 5
Progressive Disclosure	Well-organized into clearly headed sections with one-level-deep references to ../shared-references/*.md signaled via markdown links; no bundle files exist to split, and most content is appropriately placed, but it remains a single monolithic file with inline prompt templates that could live in a reference.	4 / 5
	Total	15 / 20 Passed

Description

86%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 what the skill does and gives explicit, natural trigger phrases. The main weakness is specificity, since it describes one high-level action rather than a list of concrete capabilities.

Suggestions

Enumerate 2-3 concrete deliverables the review produces (e.g., claims matrix, experiment plan, mock NeurIPS review) to lift specificity from 3 toward 4-5.

Tighten the trigger phrasing so "help me review" is scoped to research contexts to reduce overlap with generic review skills.

Dimension	Reasoning	Score
Specificity	Names the research-review domain and one concrete action ("Get a deep critical review of research") but does not enumerate multiple distinct concrete actions; closer to the 3 anchor than 4 which expects several specific actions.	3 / 5
Completeness	Explicitly answers both what ("Get a deep critical review of research from GPT using a secondary Codex agent") and when ("Use when user says..."), matching the 5 anchor with concrete trigger phrases.	5 / 5
Trigger Term Quality	Directly quotes natural user phrases ("review my research", "help me review", "get external review") plus synonyms ("critical feedback") and object types (research ideas, papers, experimental results), giving comprehensive trigger coverage.	5 / 5
Distinctiveness Conflict Risk	A clear research-review niche with specific triggers, but phrasing like "help me review" could overlap with general code/document review skills, so it is mostly distinct rather than fully distinct.	4 / 5
	Total	17 / 20 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
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/skills-codex/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.