Token导航 LogoToken导航

Agent Skill · 研究检索

solidity-code-review(可靠性代码审查)

solidity-code-review 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。

solidity-code-review 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,权限为调用外部服务;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 55,093。

总安装 490周安装 20GitHub Stars 公开资料未说明下载 157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:solidity-code-review(可靠性代码审查)
来源仓库:https://github.com/whackur/solidity-agent-toolkit
仓库路径:skills/solidity-code-review
安装命令:
npx skills add https://github.com/whackur/solidity-agent-toolkit --skill solidity-code-review
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/whackur/solidity-agent-toolkit --skill solidity-code-review

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。

solidity-code-review 用于查找、检索和筛选相关信息,支持基于关键词或任务场景定位内容。适用于需要快速获取候选结果的 Codex、Claude、Cursor 等宿主环境。通过 npx skills add 命令从指定 GitHub 仓库安装,建议结合原始 README 验证具体用法。

安装前应确认权限范围和维护状态,避免触发联网或文件操作等敏感行为。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/solidity-code-review

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/whackur/solidity-agent-toolkit --skill solidit…;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

    它需要调用外部服务,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

solidity-code-review 怎么安装?

在终端执行 npx skills add https://github.com/whackur/solidity-agent-toolkit --skill solidity-code-rev…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

solidity-code-review 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。

solidity-code-review 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

solidity-code-review 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。

solidity-code-review 用的人多吗?

公开数据是累计安装 490 次,近一周安装 20 次,下载 157 次。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。

solidity-code-review 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。

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。

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
solidity-code-review

可靠性代码审查 · 本页

490—调用外部服务未明确通过

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

Solidity Code Review Guide

When to Apply

Apply this methodology when performing a security audit, peer review, or general assessment of Solidity smart contracts. It is designed to identify vulnerabilities, ensure adherence to best practices, and verify the robustness of the contract logic.

Pre-Review Checklist

Before beginning the manual review, ensure the following items are addressed:

  • Compilation: Verify the code compiles without errors using the project's build system (Foundry, Hardhat, etc.).
  • Test Suite: Run the existing test suite. Ensure tests pass and review coverage reports to identify untested logic.
  • Dependencies: Identify all external libraries and inherited contracts. Verify versions are pinned and trusted.
  • Documentation: Review technical specifications and NatSpec comments to understand intended behavior.
  • Known Issues: Check for previous audit reports or documented "known risks" provided by the developers.
  • Scope: Define the exact list of contracts and functions that are within the audit scope.

Review Methodology

  1. Step 1: Scope & Architecture: Map out the contract inheritance, external dependencies, and system architecture.
  2. Step 2: Manual Line-by-Line Review: Perform a deep dive into critical functions, focusing on state changes and value transfers.
  3. Step 3: Automated Analysis: Run static analysis tools (Slither, Aderyn, Solhint) to catch common patterns and style violations.
  4. Step 4: Vulnerability Pattern Matching: Specifically check for known SCWE patterns (Reentrancy, Access Control, etc.).
  5. Step 5: Integration & Edge Cases: Analyze how contracts interact and test boundary conditions (e.g., zero values, max integers).

Severity Classification

SeverityCriteriaExamples
CriticalDirect loss of funds, permanent contract lock, or total compromise.Reentrancy, Unprotected withdraw, Logic error in transfer.
HighSignificant impact on system functionality or exploitable under realistic conditions.Access control bypass, Unchecked external calls, Oracle manipulation.
MediumLimited impact or requires specific, difficult-to-achieve conditions.Timestamp dependence, Front-running, Denial of Service (DoS).
LowBest practice violations, informational findings, or minor optimizations.Missing events, Floating pragma, Unused variables.

Key Inspection Areas

Access Control & Authorization

  • Verify onlyOwner or role-based access on all sensitive state-changing functions.
  • Ensure initializers are protected and can only be called once.
  • Check for tx.origin usage instead of msg.sender.

External Call Safety

  • Follow the Check-Effects-Interactions (CEI) pattern strictly.
  • Use call() instead of transfer() or send() for ETH transfers.
  • Handle return values of all external calls.

State Management & Reentrancy

  • Use ReentrancyGuard for functions making external calls.
  • Check for cross-contract reentrancy where state is shared.
  • Ensure state variables are updated before external interactions.

Arithmetic & Type Safety

  • For Solidity <0.8.0, ensure SafeMath is used.
  • Check for precision loss in divisions (multiply before dividing).
  • Verify safe casting between types (e.g., uint256 to uint8).

Token Handling (ERC20/721)

  • Use SafeERC20 for transfer and transferFrom.
  • Account for "fee-on-transfer" tokens if applicable.
  • Verify approve race condition handling.

Upgrade Mechanisms

  • Check for storage gaps in logic contracts to prevent collisions.
  • Ensure logic contracts do not use selfdestruct or delegatecall.
  • Verify the proxy admin has restricted access.

Event Emissions

  • Emit events for all significant state changes (ownership, parameters, transfers).
  • Use indexed parameters for efficient off-chain filtering.

NatSpec Documentation

  • Ensure @notice, @param, and @return are accurate.
  • Use @dev to document complex logic or security assumptions.

Style Guide Compliance

  • Code follows the official Solidity style guide conventions.
  • Naming conventions: PascalCase (contracts), camelCase (functions), UPPER_CASE (constants).
  • Function ordering: by visibility (external → public → internal → private), then by mutability (state-changing → view → pure) within each group.
  • Function modifier order: visibility, mutability, virtual, override, custom.
  • See the Solidity Style Guide Reference for the full checklist.

Reporting Format

Findings should be documented using the following template:

[SEVERITY] Finding Title

ID: SCWE-XXX *(replace with actual SCWE ID, e.g., SCWE-046 — see search_vulnerabilities)* Location: ContractName.sol:L42 Description: Detailed explanation of the vulnerability and how it can be triggered. Impact: What happens if this is exploited (e.g., "User funds can be stolen"). Remediation: Specific code changes or architectural adjustments to fix the issue.

Enhanced with MCP

When using the solidity-agent-toolkit, leverage these tools in a structured review workflow:

Step 1 — Static Analysis:

  • run_slither — Comprehensive static analysis with SCWE-mapped findings
  • run_aderyn — Fast Rust-based vulnerability scanner
  • run_solhint — Linting and style enforcement

Step 2 — Pattern Detection:

  • match_vulnerability_patterns — Regex-based detection of 32+ common vulnerability patterns

Step 3 — Vulnerability Lookup:

  • search_vulnerabilities — Search the OWASP SCWE database by keyword
  • get_remediation — Get specific fix guidance with code examples for any SCWE ID
  • check_vulnerability — Check if code matches a known SCWE pattern

Step 4 — Style & Quality:

  • check_style — Automated Solidity style guide compliance (12 rules)
  • format_code — Auto-format Solidity code
  • validate_natspec — Verify NatSpec documentation completeness

Step 5 — Full Audit:

  • Use the security_audit prompt for a structured, guided audit process
  • Use the code_review prompt for comprehensive code quality assessment

References

平台分布

Codex

34.05%
按下载量换算53

Claude

27.81%
按下载量换算44

Cursor

20.26%
按下载量换算32

Gemini CLI

10.1%
按下载量换算16

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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