review

Use when a draft, paper, or paper-like report is substantial enough for an independent skeptical audit before finalization, rebuttal, or revision routing.

Quality

76%

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

Fix and improve this skill with Tessl

tessl review fix ./src/skills/review/SKILL.md

Quality

Content

77%

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

The body is highly actionable with a clear, gated multi-step workflow and real referenced templates, but its length and redundancy across sections, plus inline content that could live in references, hold back conciseness and progressive disclosure.

Suggestions

Deduplicate the review_followup_policy guidance so it is stated once and referenced elsewhere, removing the repetition between Non-negotiable rules and Workflow section 7.

Move the detailed experiment-matrix row schema and the long review-dimensions list into a reference file, keeping SKILL.md as a lean overview that links to it.

Tighten the repeated 'Do not ...' rule clusters into a single compact list to recover token budget.

Dimension	Reasoning	Score
Conciseness	The ~500-line body is mostly domain-specific procedural content rather than concepts Claude already knows, but it restates the same rules in multiple places (e.g., review_followup_policy appears in Non-negotiable rules and again in Workflow section 7) and could be materially tightened.	2 / 3
Actionability	For an instruction-only skill the guidance is concrete and executable: exact artifact paths, explicit tool calls like `artifact.validate_manuscript_coverage(detail='full')` and `memory.list_recent(scope='quest', limit=5)`, and a full matrix-row schema with named fields.	3 / 3
Workflow Clarity	An explicit 8-step workflow (plan audit, literature benchmark, write review, revision log, experiment TODO, route, auto follow-up, manuscript revision) is paired with validation gates such as the evidence-authenticity audit and 'before treating experiments as stable, require ...'.	3 / 3
Progressive Disclosure	Templates are correctly offloaded to one-level-deep, clearly signaled references/ files that exist on disk, but large inline blocks (the matrix row schema, the full review-dimensions list, repeated policy text) could themselves be split into referenced files.	2 / 3
	Total	10 / 12 Passed

Description

75%

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 well-structured with a clear 'Use when' trigger and a distinct, non-conflicting niche, but its single-action phrasing and limited keyword spread keep specificity and trigger-term quality at the mid level.

Suggestions

Add a few more natural trigger phrases (e.g., 'submission', 'review my paper', 'is my draft good enough') to broaden trigger-term coverage.

Name a second concrete action (e.g., 'write a revision plan' or 'produce a follow-up experiment TODO list') to reach the multi-action specificity anchor.

Dimension	Reasoning	Score
Specificity	Names a concrete domain (paper/draft/report) and one core action ("independent skeptical audit") but does not list multiple specific actions, stopping short of the multi-action score-3 anchor.	2 / 3
Completeness	The "Use when ..." clause explicitly states the trigger (a substantial draft/paper/report needing audit before finalization/rebuttal/revision) while the action (independent skeptical audit) answers both what and when.	3 / 3
Trigger Term Quality	Includes natural terms like "draft", "paper", and "paper-like report" that users would say, but omits common variations such as "submission" or "review my paper" that would give full coverage.	2 / 3
Distinctiveness Conflict Risk	The audit niche plus explicit routing to finalization/rebuttal/revision gives a distinct trigger that is unlikely to fire for unrelated skills, and the body further separates it from `write` and `rebuttal`.	3 / 3
	Total	10 / 12 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
skill_md_line_count	SKILL.md is long (507 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: ResearAI/DeepScientist
Path: src/skills/review/SKILL.md
Commit: b366244

Reviewed: 2 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.