auto-review-loop-minimax

Autonomous multi-round research review loop using MiniMax API. Use when you want to use MiniMax instead of Codex MCP for external review. Trigger with "auto review loop minimax" or "minimax review".

Quality

81%

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

81%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 highly actionable and workflow-clear, with concrete code, an explicit stop condition, and state-recovery logic. Its main weakness is conciseness: the curl fallback and system prompt are duplicated verbatim across sections rather than defined once and referenced.

Suggestions

Define the curl review-call template and the reviewer system prompt once (e.g., in a short constants block or a referenced snippet) and reuse it in both Phase A and the Round 2+ template to remove the verbatim duplication.

Consolidate the MCP-vs-curl branching so the same prompt body is not restated for each transport; reference a single prompt structure instead.

Consider moving the long Round 2+ curl block into a references file (e.g., references/round2-prompt.md) and linking to it, improving both conciseness and progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Content is operational throughout with no concept-padding Claude already knows, but the full curl fallback block and system prompt are duplicated verbatim in both Phase A and the Round 2+ template, which is unnecessary padding that could be tightened.	3 / 5
Actionability	Provides fully executable, copy-paste-ready guidance on both paths (concrete MCP tool invocation and complete curl commands with real endpoint, headers, model, and max_tokens), plus a concrete JSON state schema and markdown log template.	5 / 5
Workflow Clarity	Phases A-E are clearly sequenced with an explicit STOP CONDITION, fresh-start/resume branching in Initialization, and a state-persistence checkpoint each round, giving a clear feedback loop (review -> fix -> re-review) with explicit validation gates.	5 / 5
Progressive Disclosure	Well-organized sections with clearly signaled one-level-deep references to shared-references/*.md, but the large prompt templates (full curl blocks) are inlined twice rather than factored out, leaving a minor organization gap.	4 / 5
	Total	17 / 20 Passed

Description

82%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 answers both what and when with explicit trigger phrases and is clearly distinguishable from its Codex-based sibling. Its only gap is that it does not enumerate several specific concrete actions, capping specificity at 3.

Dimension	Reasoning	Score
Specificity	Names the domain ('Autonomous multi-round research review loop') and a concrete mechanism ('using MiniMax API') but does not enumerate several specific concrete actions, matching the anchor that lists domain plus 1-2 actions.	3 / 5
Completeness	Clearly states what the skill does ('Autonomous multi-round research review loop using MiniMax API') and when to use it ('Use when you want to use MiniMax instead of Codex MCP for external review') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Provides explicit natural trigger phrases users would say ('auto review loop minimax', 'minimax review') with good keyword coverage, though not a comprehensive set of synonyms or file extensions.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche as the MiniMax-specific variant, explicitly disambiguates from the Codex MCP sibling ('instead of Codex MCP'), and uses distinct trigger phrases, minimizing conflict risk.	5 / 5
	Total	17 / 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: 4 suspicious	Warning

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/auto-review-loop-minimax/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.