Token导航 LogoToken导航

Agent Skill · 研究检索

multiversx-spec-compliance(multiversx 规范合规性)

multiversx-spec-compliance 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。

multiversx-spec-compliance 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;许可证信息还不明确,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 54,679。

总安装 499周安装 20GitHub Stars 11下载 162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
unknown

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:multiversx-spec-compliance(multiversx 规范合规性)
来源仓库:https://github.com/multiversx/mx-ai-skills
仓库路径:skills/multiversx-spec-compliance
安装命令:
npx skills add https://github.com/multiversx/mx-ai-skills --skill multiversx-spec-compliance
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/multiversx/mx-ai-skills --skill multiversx-spec-compliance

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。

用于查找、检索和筛选相关信息,验证是否符合 MultiversX 规范。适合在开发或审计时对照技术标准进行检查。可结合官方文档生成合规性报告,但需人工确认例外情况。

安装命令:npx skills add https://github.com/multiversx/mx-ai-skills --skill multiversx-spec-compliance。注意规范更新频率,避免使用过时的标准版本。

来源仓库内路径:skills/multiversx-spec-compliance

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/multiversx/mx-ai-skills --skill multiversx-spe…;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

multiversx-spec-compliance 怎么安装?

在终端执行 npx skills add https://github.com/multiversx/mx-ai-skills --skill multiversx-spec-complian…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

multiversx-spec-compliance 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。

multiversx-spec-compliance 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。

multiversx-spec-compliance 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

multiversx-spec-compliance 用的人多吗?

公开数据是累计安装 499 次,近一周安装 20 次,下载 162 次,GitHub 11 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。

multiversx-spec-compliance 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。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
multiversx-spec-compliance

multiversx 规范合规性 · 本页

49911只读未明确通过

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

Specification Compliance Verification

Ensure that MultiversX smart contract implementations match their intended design as specified in whitepapers, technical specifications, and MultiversX Improvement Proposals (MIPs). This skill bridges the gap between documentation and code.

When to Use

  • Auditing contracts against their whitepapers
  • Verifying tokenomics implementations
  • Checking MIP standard compliance
  • Validating economic formulas and constraints
  • Reviewing upgrade proposals against specs

1. Verification Process Overview

Inputs Required

InputDescriptionSource
CodeRust implementationsrc/*.rs
SpecificationDesign documentwhitepaper.pdf, README.md, specs/
MIP ReferenceStandard requirementsMultiversX MIPs

Process Flow

1. Extract Claims → List all requirements from spec
2. Map to Code   → Find implementing code for each claim
3. Verify Logic  → Confirm implementation matches spec
4. Document      → Record findings and deviations

2. Claim Extraction

Specification Language Keywords

Extract statements containing these keywords:

KeywordMeaningExample
MUSTRequired"Users MUST stake minimum 100 tokens"
MUST NOTForbidden"Admin MUST NOT withdraw user funds"
SHOULDRecommended"Contract SHOULD emit events"
SHALLObligation"Rewards SHALL be calculated daily"
MAYOptional"Users MAY delegate to multiple validators"

Example Claim Extraction

From Whitepaper:

"The staking contract MUST enforce a minimum stake of 1000 EGLD. Rewards MUST be calculated using APY = base_rate * (1 + boost_factor). Users MUST NOT be able to withdraw during the lock period."

Extracted Claims:

1. [MUST] Minimum stake: 1000 EGLD
2. [MUST] Reward formula: APY = base_rate * (1 + boost_factor)
3. [MUST NOT] Withdrawal during lock period

Claim Documentation Template

| ID | Type | Claim | Source | Code Location | Status |
|----|------|-------|--------|---------------|--------|
| C1 | MUST | Min stake 1000 EGLD | WP §3.1 | stake.rs:45 | Verified |
| C2 | MUST | APY formula | WP §4.2 | rewards.rs:78 | Deviation |
| C3 | MUST NOT | Lock withdrawal | WP §3.3 | withdraw.rs:23 | Verified |

3. Code Mapping

Finding Implementing Code

For each claim, locate the relevant code:

// Claim C1: Min stake 1000 EGLD
// Location: src/stake.rs:45

const MIN_STAKE_EGLD: u64 = 1000;  // 1000 EGLD (whole units)
const DECIMALS: u32 = 18;

#[payable("EGLD")]
#[endpoint]
fn stake(&self) {
    let payment = self.call_value().egld();
    let min_stake_wei = BigUint::from(MIN_STAKE_EGLD) * BigUint::from(10u64).pow(DECIMALS);
    require!(
        *payment >= min_stake_wei,
        "Minimum stake is 1000 EGLD"  // ← Implements C1
    );
    // ...
}

Mapping Checklist

For each claim:

  • Code location identified
  • Implementation logic understood
  • Constants/values match spec
  • Edge cases handled per spec

4. Verification Techniques

Formula Verification

Spec:

"APY = base_rate * (1 + boost_factor)"

Code Review:

fn calculate_apy(&self, base_rate: BigUint, boost_factor: BigUint) -> BigUint {
    // Verify this matches: APY = base_rate * (1 + boost_factor)

    let one = BigUint::from(PRECISION);  // Check: What is PRECISION?
    let boost_multiplier = &one + &boost_factor;
    let apy = &base_rate * &boost_multiplier / &one;

    // QUESTION: Is division by PRECISION correct? Spec doesn't mention it.
    // FINDING: Precision handling not in spec - potential deviation

    apy
}

Constraint Verification

Spec:

"Users MUST NOT withdraw during the lock period of 7 days"

Code Review:

#[endpoint]
fn withdraw(&self) {
    let stake_time = self.stake_timestamp(&caller).get(); // TimestampMillis
    let current_time = self.blockchain().get_block_timestamp_millis();
    let lock_period = self.lock_period().get();  // DurationMillis - Check: Is this 7 days?

    require!(
        current_time >= stake_time + lock_period,
        "Lock period not elapsed"
    );
    // ...
}

// VERIFICATION NEEDED:
// 1. Is lock_period initialized to 7 days (604800 seconds)?
// 2. Is lock_period immutable or can admin change it?
// 3. Can this be bypassed through any other endpoint?

State Transition Verification

Spec:

"State transitions: INACTIVE → ACTIVE → COMPLETED"

Code Review:

#[derive(TopEncode, TopDecode, TypeAbi, PartialEq)]
pub enum State {
    Inactive,
    Active,
    Completed,
}

fn activate(&self) {
    let current = self.state().get();
    require!(current == State::Inactive, "Can only activate from Inactive");
    self.state().set(State::Active);
}

fn complete(&self) {
    let current = self.state().get();
    require!(current == State::Active, "Can only complete from Active");
    self.state().set(State::Completed);
}

// VERIFICATION:
// ✓ Inactive → Active (activate)
// ✓ Active → Completed (complete)
// ? Is there a way to go backwards? (Should not be allowed)
// ? Can state be set directly? (Search for .set(State::))

5. MultiversX MIP Compliance

Common MIPs to Verify

MIPTopicKey Requirements
MIP-2Semi-Fungible TokensSFT metadata format, royalties
MIP-3Dynamic NFTsAttribute update mechanisms
MIP-4RoyaltiesRoyalty calculation and distribution

MIP-2 SFT Compliance Example

Requirements:

  • Token type must be SFT (nonce > 0, quantity > 1 allowed)
  • Metadata format follows standard
  • Royalties encoded correctly

Verification:

// Check NFT creation follows MIP-2

#[endpoint]
fn create_sft(&self, ...) -> u64 {
    // VERIFY: Using NonFungibleTokenMapper correctly
    let nonce = self.sft_token().nft_create(
        initial_quantity,  // MIP-2: Must allow quantity > 1
        &SftAttributes {
            // MIP-2: Required attributes
            name: ...,
            royalties: ...,  // In basis points (0-10000)
            hash: ...,
            attributes: ...,
            uris: ...,
        }
    );
    nonce
}

6. Tokenomics Verification

Common Tokenomics Claims

Claim TypeExampleVerification
Total SupplyMax 1B tokensCheck mint constraints
Inflation Rate5% annuallyVerify mint formula
Burn Rate1% per transferCheck fee calculation
Distribution40% communityVerify initial allocation

Example: Inflation Verification

Spec:

"Annual inflation rate is 5%, calculated per epoch"

Code Review:

const ANNUAL_INFLATION_BPS: u64 = 500;  // 5% = 500 basis points
const EPOCHS_PER_YEAR: u64 = 365;       // Assuming daily epochs

fn calculate_epoch_inflation(&self) -> BigUint {
    let total_supply = self.total_supply().get();
    let epoch_rate = ANNUAL_INFLATION_BPS / EPOCHS_PER_YEAR;

    // VERIFICATION:
    // 500 / 365 = 1.369... but integer division = 1
    // This is LESS than 5% annually (365 * 1 = 365 bps = 3.65%)
    // FINDING: Integer precision loss causes ~27% less inflation than spec

    &total_supply * BigUint::from(epoch_rate) / BigUint::from(10000u64)
}

7. Deviation Handling

Deviation Categories

CategorySeverityAction
CriticalBreaks core functionalityMust fix
MajorSignificant differenceShould fix
MinorSlight variationDocument
EnhancementBeyond specDocument

Deviation Report Template

## Deviation Report

### DEV-001: Inflation Calculation Precision Loss

**Claim**: Annual inflation rate is 5%
**Source**: Whitepaper §5.2
**Code**: rewards.rs:calculate_epoch_inflation()

**Expected**: 5.00% annual inflation
**Actual**: 3.65% annual inflation

**Root Cause**: Integer division of basis points by epochs
loses precision (500/365 = 1, not 1.369)

**Impact**: ~27% less inflation than documented

**Recommendation**: Use scaled arithmetic

// Instead of: let epoch_rate = ANNUAL_INFLATION_BPS / EPOCHS_PER_YEAR;

// Use: let scaled_annual = BigUint::from(ANNUAL_INFLATION_BPS) * &total_supply; let epoch_inflation = scaled_annual / BigUint::from(EPOCHS_PER_YEAR) / BigUint::from(10000u64);


**Severity**: Major **Status**: Open

8. Compliance Report Template

# Specification Compliance Report

**Project**: [Name]
**Specification Version**: [Version]
**Code Version**: [Commit/Tag]
**Date**: [Date]
**Auditor**: [Name]

## Executive Summary
[Brief overview of compliance status]

## Specification Coverage

| Section | Claims | Verified | Deviations | Not Found |
|---------|--------|----------|------------|-----------|
| §3 Staking | 12 | 10 | 1 | 1 |
| §4 Rewards | 8 | 7 | 1 | 0 |
| §5 Governance | 5 | 5 | 0 | 0 |
| **Total** | **25** | **22** | **2** | **1** |

## Verified Claims
[List of all verified claims with code references]

## Deviations
[Detailed deviation reports]

## Unimplemented Claims
[Claims from spec not found in code]

## MIP Compliance
| MIP | Status | Notes |
|-----|--------|-------|
| MIP-2 | Compliant | - |
| MIP-4 | Partial | Royalty distribution differs |

## Recommendations
1. [Priority recommendation]
2. [Second priority]

## Conclusion
[Overall compliance assessment]

9. Best Practices

  1. Get the right spec version: Ensure code and spec versions match
  2. Document assumptions: When spec is ambiguous, document interpretation
  3. Test boundary values: Verify spec limits are correctly implemented
  4. Check units: EGLD vs wei, seconds vs epochs, basis points vs percentages
  5. Verify precision: BigUint calculations should maintain precision
  6. Review change history: Check if spec evolved and code was updated

平台分布

Codex

35.53%
按下载量换算58

Claude

28.7%
按下载量换算46

Cursor

18.33%
按下载量换算30

Gemini CLI

10.5%
按下载量换算17

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills