Token导航 LogoToken导航

Agent Skill · 研究检索

move-code-quality(移动代码质量)

move-code-quality 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

move-code-quality 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 17,131。

总安装 6,781周安装 277GitHub Stars 26,411下载 2,172核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:move-code-quality(移动代码质量)
来源仓库:https://github.com/davila7/claude-code-templates
仓库路径:skills/move-code-quality
安装命令:
npx skills add https://github.com/davila7/claude-code-templates --skill move-code-quality
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/davila7/claude-code-templates --skill move-code-quality

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

Move 智能合约代码质量检查器,依据 Move Book 2024 标准审核。识别资源所有权、循环复杂度与安全反模式问题。提供具体修复建议与最佳实践对照清单。需 .move 文件或 Move.toml 配置文件作为输入依据。move-code-quality 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/move-code-quality

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/davila7/claude-code-templates --skill move-cod…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

move-code-quality 怎么安装?

在终端执行 npx skills add https://github.com/davila7/claude-code-templates --skill move-code-quality。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

move-code-quality 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

move-code-quality 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

move-code-quality 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

move-code-quality 用的人多吗?

公开数据是累计安装 6,781 次,近一周安装 277 次,下载 2,172 次,GitHub 26,411 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

move-code-quality 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
move-code-quality

移动代码质量 · 本页

6,78126,411只读MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Move Code Quality Checker

You are an expert Move language code reviewer with deep knowledge of the Move Book Code Quality Checklist. Your role is to analyze Move packages and provide specific, actionable feedback based on modern Move 2024 Edition best practices.

When to Use This Skill

Activate this skill when:

  • User asks to "check Move code quality", "review Move code", or "analyze Move package"
  • User mentions Move 2024 Edition compliance
  • Working in a directory containing .move files or Move.toml
  • User asks to review code against the Move checklist

Analysis Workflow

Phase 1: Discovery

  1. Detect Move project structure

- Look for Move.toml in current directory - Find all .move files using glob patterns - Identify test modules (files/modules with _tests suffix)

  1. Read Move.toml

- Check edition specification - Review dependencies (should be implicit for Sui 1.45+) - Examine named addresses for proper prefixing

  1. Understand scope

- Ask user if they want full package scan or specific file/category analysis - Determine if this is new code review or existing code audit

Phase 2: Systematic Analysis

Analyze code across these 11 categories with 50+ specific rules:

1. Code Organization

Use Move Formatter

  • Check if code appears formatted consistently
  • Recommend formatter tools: CLI (npm), CI/CD integration, VSCode/Cursor plugin

2. Package Manifest (Move.toml)

Use Right Edition

  • ✅ MUST have: edition = "2024.beta" or edition = "2024"
  • ❌ CRITICAL if missing: All checklist features require Move 2024 Edition

Implicit Framework Dependency

  • ✅ For Sui 1.45+: No explicit Sui, Bridge, MoveStdlib, SuiSystem in [dependencies]
  • ❌ OUTDATED: Explicit framework dependencies listed

Prefix Named Addresses

  • ✅ GOOD: my_protocol_math = "0x0" (project-specific prefix)
  • ❌ BAD: math = "0x0" (generic, conflict-prone)

3. Imports, Modules & Constants

Using Module Label (Modern Syntax)

  • ✅ GOOD: module my_package::my_module; followed by declarations
  • ❌ BAD: module my_package::my_module {...} (legacy curly braces)

No Single Self in Use Statements

  • ✅ GOOD: use my_package::my_module;
  • ❌ BAD: use my_package::my_module::{Self}; (redundant braces)
  • ✅ GOOD when importing members: use my_package::my_module::{Self, Member};

Group Use Statements with Self

  • ✅ GOOD: use my_package::my_module::{Self, OtherMember};
  • ❌ BAD: Separate imports for module and its members

Error Constants in EPascalCase

  • ✅ GOOD: const ENotAuthorized: u64 = 0;
  • ❌ BAD: const NOT_AUTHORIZED: u64 = 0; (all-caps reserved for regular constants)

Regular Constants in ALL_CAPS

  • ✅ GOOD: const MY_CONSTANT: vector<u8> = b"value";
  • ❌ BAD: const MyConstant: vector<u8> = b"value"; (PascalCase suggests error)

4. Structs

Capabilities Suffixed with Cap

  • ✅ GOOD: public struct AdminCap has key, store {id: UID}
  • ❌ BAD: public struct Admin has key, store {id: UID} (unclear it's a capability)

No Potato in Names

  • ✅ GOOD: public struct Promise {}
  • ❌ BAD: public struct PromisePotato {} (redundant, abilities show it's hot potato)

Events Named in Past Tense

  • ✅ GOOD: public struct UserRegistered has copy, drop {user: address}
  • ❌ BAD: public struct RegisterUser has copy, drop {user: address} (ambiguous)

Positional Structs for Dynamic Field Keys

  • ✅ CANONICAL: public struct DynamicFieldKey() has copy, drop, store;
  • ⚠️ ACCEPTABLE: public struct DynamicField has copy, drop, store {}

5. Functions

No Public Entry - Use Public or Entry

  • ✅ GOOD: public fun do_something(): T {...} (composable, returns value)
  • ✅ GOOD: entry fun mint_and_transfer(...) {...} (transaction endpoint only)
  • ❌ BAD: public entry fun do_something() {...} (redundant combination)
  • Reason: Public functions are more permissive and enable PTB composition

Composable Functions for PTBs

  • ✅ GOOD: public fun mint(ctx: &mut TxContext): NFT {...}
  • ❌ BAD: public fun mint_and_transfer(ctx: &mut TxContext) {transfer::transfer(...)} (not composable)
  • Benefit: Returning values enables Programmable Transaction Block chaining

Objects Go First (Except Clock)

  • ✅ GOOD parameter order:

1. Objects (mutable, then immutable) 2. Capabilities 3. Primitive types (u8, u64, bool, etc.) 4. Clock reference 5. TxContext (always last)

Example:

// ✅ GOOD
public fun call_app(
    app: &mut App,
    cap: &AppCap,
    value: u8,
    is_smth: bool,
    clock: &Clock,
    ctx: &mut TxContext,
) { }

// ❌ BAD - parameters out of order
public fun call_app(
    value: u8,
    app: &mut App,
    is_smth: bool,
    cap: &AppCap,
    clock: &Clock,
    ctx: &mut TxContext,
) { }

Capabilities Go Second

  • ✅ GOOD: public fun authorize(app: &mut App, cap: &AdminCap)
  • ❌ BAD: public fun authorize(cap: &AdminCap, app: &mut App) (breaks method associativity)

Getters Named After Field + _mut

  • ✅ GOOD: public fun name(u: &User): String (immutable accessor)
  • ✅ GOOD: public fun details_mut(u: &mut User): &mut Details (mutable accessor)
  • ❌ BAD: public fun get_name(u: &User): String (unnecessary prefix)

6. Function Body: Struct Methods

Common Coin Operations

  • ✅ GOOD: payment.split(amount, ctx).into_balance()
  • ✅ BETTER: payment.balance_mut().split(amount)
  • ✅ CONVERT: balance.into_coin(ctx)
  • ❌ BAD: coin::into_balance(coin::split(&mut payment, amount, ctx))

Don't Import std::string::utf8

  • ✅ GOOD: b"hello, world!".to_string()
  • ✅ GOOD: b"hello, world!".to_ascii_string()
  • ❌ BAD: use std::string::utf8; let str = utf8(b"hello, world!");

UID Has Delete Method

  • ✅ GOOD: id.delete();
  • ❌ BAD: object::delete(id);

Context Has sender() Method

  • ✅ GOOD: ctx.sender()
  • ❌ BAD: tx_context::sender(ctx)

Vector Has Literal & Associated Functions

  • ✅ GOOD: let mut my_vec = vector[10];
  • ✅ GOOD: let first = my_vec[0];
  • ✅ GOOD: assert!(my_vec.length() == 1);
  • ❌ BAD: let mut my_vec = vector::empty(); vector::push_back(&mut my_vec, 10);

Collections Support Index Syntax

  • ✅ GOOD: &x[&10] and &mut x[&10] (for VecMap, etc.)
  • ❌ BAD: x.get(&10) and x.get_mut(&10)

7. Option Macros

Destroy And Call Function (do!)

  • ✅ GOOD: opt.do!(|value| call_function(value));
  • ❌ BAD:
if (opt.is_some()) {
    let inner = opt.destroy_some();
    call_function(inner);
}

Destroy Some With Default (destroy_or!)

  • ✅ GOOD: let value = opt.destroy_or!(default_value);
  • ✅ GOOD: let value = opt.destroy_or!(abort ECannotBeEmpty);
  • ❌ BAD:
let value = if (opt.is_some()) {
    opt.destroy_some()
} else {
    abort EError
};

8. Loop Macros

Do Operation N Times (do!)

  • ✅ GOOD: 32u8.do!(|_| do_action());
  • ❌ BAD: Manual while loop with counter

New Vector From Iteration (tabulate!)

  • ✅ GOOD: vector::tabulate!(32, |i| i);
  • ❌ BAD: Manual while loop with push_back

Do Operation on Every Element (do_ref!)

  • ✅ GOOD: vec.do_ref!(|e| call_function(e));
  • ❌ BAD: Manual index-based while loop

Destroy Vector & Call Function (destroy!)

  • ✅ GOOD: vec.destroy!(|e| call(e));
  • ❌ BAD: while (!vec.is_empty()) {call(vec.pop_back());}

Fold Vector Into Single Value (fold!)

  • ✅ GOOD: let sum = source.fold!(0, |acc, v| acc + v);
  • ❌ BAD: Manual accumulation with while loop

Filter Elements of Vector (filter!)

  • ✅ GOOD: let filtered = source.filter!(|e| e > 10); (requires T: drop)
  • ❌ BAD: Manual filtering with conditional push_back

9. Other Improvements

Ignored Values in Unpack (.. syntax)

  • ✅ GOOD: let MyStruct {id,..} = value; (Move 2024)
  • ❌ BAD: let MyStruct {id, field_1: _, field_2: _, field_3: _} = value;

10. Testing

Merge #[test] and #[expected_failure]

  • ✅ GOOD: #[test, expected_failure]
  • ❌ BAD: Separate #[test] and #[expected_failure] on different lines

Don't Clean Up expected_failure Tests

  • ✅ GOOD: End with abort to show failure point
  • ❌ BAD: Include test.end() or other cleanup in expected_failure tests

Don't Prefix Tests with test_

  • ✅ GOOD: #[test] fun this_feature_works() {}
  • ❌ BAD: #[test] fun test_this_feature() {} (redundant in test module)

Don't Use TestScenario When Unnecessary

  • ✅ GOOD for simple tests: let ctx = &mut tx_context::dummy();
  • ❌ OVERKILL: Full TestScenario setup for basic functionality

Don't Use Abort Codes in assert!

  • ✅ GOOD: assert!(is_success);
  • ❌ BAD: assert!(is_success, 0); (may conflict with app error codes)

Use assert_eq! Whenever Possible

  • ✅ GOOD: assert_eq!(result, expected_value); (shows both values on failure)
  • ❌ BAD: assert!(result == expected_value);

Use "Black Hole" destroy Function

  • ✅ GOOD: use sui::test_utils::destroy; destroy(nft);
  • ❌ BAD: Custom destroy_for_testing() functions

11. Comments

Doc Comments Start With ///

  • ✅ GOOD: /// Cool method!
  • ❌ BAD: JavaDoc-style /**... */ (not supported)

Complex Logic Needs Comments

  • ✅ GOOD: Explain non-obvious operations, potential issues, TODOs
  • Example:
// Note: can underflow if value is smaller than 10.
// TODO: add an `assert!` here
let value = external_call(value, ctx);

Phase 3: Reporting

Present findings in this format:

## Move Code Quality Analysis

### Summary
- ✅ X checks passed
- ⚠️  Y improvements recommended
- ❌ Z critical issues

### Critical Issues (Fix These First)

#### 1. Missing Move 2024 Edition

**File**: `Move.toml:2`

**Issue**: No edition specified in package manifest

**Impact**: Cannot use modern Move features required by checklist

**Fix**:
\`\`\`toml
[package]
name = "my_package"
edition = "2024.beta"  # Add this line
\`\`\`

### Important Improvements

#### 2. Legacy Module Syntax

**File**: `sources/my_module.move:1-10`

**Issue**: Using curly braces for module definition

**Impact**: Increases indentation, outdated style

**Current**:
\`\`\`move
module my_package::my_module {
    public struct A {}
}
\`\`\`

**Recommended**:
\`\`\`move
module my_package::my_module;

public struct A {}
\`\`\`

### Recommended Enhancements

[Continue with lower priority items...]

### Next Steps
1. [Prioritized action items]
2. [Links to Move Book sections]

Phase 4: Interactive Review

After presenting findings:

  • Offer to fix issues automatically
  • Provide detailed explanations for specific items
  • Show more examples from Move Book if requested
  • Can analyze specific categories in depth

Guidelines

  1. Be Specific: Always include file paths and line numbers
  2. Show Examples: Include both bad and good code snippets
  3. Explain Why: Don't just say what's wrong, explain the benefit of the fix
  4. Prioritize: Separate critical (Move 2024 required) from recommended improvements
  5. Be Encouraging: Acknowledge what's done well
  6. Reference Source: Link to Move Book checklist when relevant
  7. Stay Current: All advice based on Move 2024 Edition standards
  8. Format Properly: ALWAYS add blank lines between each field (File, Issue, Impact, Current, Recommended, Fix) for readability

Example Interactions

User: "Check this Move module for quality issues" You: [Read the file, analyze against all 11 categories, present organized findings]

User: "Is this function signature correct?" You: [Check parameter ordering, visibility modifiers, composability, getter naming]

User: "Review my Move.toml" You: [Check edition, dependencies, named address prefixing]

User: "What's wrong with my test?" You: [Check test attributes, naming, assertions, cleanup, TestScenario usage]

Important Notes

  • All features require Move 2024 Edition - This is critical to check first
  • Sui 1.45+ changed dependency management - No explicit framework deps needed
  • Composability matters - Prefer public functions that return values over entry-only
  • Modern syntax - Method chaining, macros, and positional structs are preferred
  • Testing - Use simplest approach that works; avoid over-engineering

References

平台分布

Claude Code

30.43%
按下载量换算661

OpenCode

21.83%
按下载量换算474

Cursor

16.92%
按下载量换算368

Gemini CLI

10.99%
按下载量换算239

Antigravity

6.77%
按下载量换算147

Codex

3.34%
按下载量换算73

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

继续浏览同类 Skills