auto-review-loop

Autonomous multi-round research review loop. Repeatedly reviews using Gemini via gemini-review MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says "auto review loop", "review until it passes", or wants autonomous iterative improvement.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

88%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 thorough, highly actionable multi-phase workflow with strong validation checkpoints and concrete MCP calls. Its main weakness is repetition — the SCOPE LIMITS block appears twice verbatim — and inlining of large prompt templates that could be factored into a reference file.

Suggestions

De-duplicate the SCOPE LIMITS block: define it once and reference it from both the Phase A and Round 2+ prompt templates, or extract it to a referenced file.

Remove self-referential meta-commentary such as "Earlier wording used 'or' + a stale verdict set; the AND form is authoritative" — state only the authoritative rule.

Factor the full Round 2+ prompt template into a reference file and keep SKILL.md focused on the workflow, reducing inline token weight.

Dimension	Reasoning	Score
Conciseness	Mostly lean and assumes Claude's competence (no priming on what MCP or Gemini is), but the SCOPE LIMITS block is duplicated verbatim across both prompt templates and there is self-referential meta-commentary ("Earlier wording used...the AND form is authoritative") that could be trimmed.	4 / 5
Actionability	Provides fully executable guidance: exact MCP tool names with parameters, polling instructions, a concrete JSON state schema, and copy-paste-ready prompt templates covering the common round-1 and round-N+ cases.	5 / 5
Workflow Clarity	Clear phase sequence (Initialization → Loop Phases A–E → Termination) with explicit validation gates — STOP CONDITION check, a verbatim raw-response save checkpoint, compaction-recovery state persistence, and a review→fix→re-review feedback loop.	5 / 5
Progressive Disclosure	Well-organized into clear sections with one-level-deep external references (shared-references/*.md), but no bundle files exist and the large duplicated SCOPE LIMITS block plus full prompt templates are inlined where a single referenced file would reduce repetition.	4 / 5
	Total	18 / 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 that crisply states the capability and supplies concrete natural-language triggers, covering both what and when. Slight room to broaden trigger synonyms and enumerate fix types, but no real conflict risk.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "reviews", "implements fixes", and "re-reviews" — in the research-review domain, with only minor coverage gaps (does not enumerate the kinds of fixes or review outputs).	4 / 5
Completeness	Explicitly answers both what ("repeatedly reviews... implements fixes, and re-reviews until positive assessment or max rounds") and when ("Use when user says \"auto review loop\", \"review until it passes\"...") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrases a user would actually say — "auto review loop" and "review until it passes" — plus "autonomous iterative improvement", but is missing common synonyms or variations a user might utter.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Gemini-via-MCP autonomous review loop) with distinct, specific triggers that are unlikely to fire for unrelated skills.	5 / 5
	Total	18 / 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: 3 suspicious	Warning

	Total	15 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/skills-codex-gemini-review/auto-review-loop/SKILL.md
Commit: 0472e53

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.