Account Risk Skill — Development skill for Claude Code
反封号.skill — Claude Code / Anthropic / OpenAI account suspension risk detection, IP cleanliness scoring, compliance preflight, and appeal-packet skill.
How to install Account Risk Skill
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open swaylq/account-risk-skill and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Account Risk Skill does
反封号.skill — Claude Code / Anthropic / OpenAI account suspension risk detection, IP cleanliness scoring, compliance preflight, and appeal-packet skill. Compliance-only; no ban evasion.
Alternatives in Development
- Privacy Policy — Draft privacy policies with GDPR compliance considerations 7.8k ★
- Contributing To N-skills — This is a curated marketplace 930 ★
- Deep Research From Scratch — Deep research has broken out as one of the most popular agent applications 768 ★
README
🛡️ 反封号.skill
*「装到 Claude Code,问一句:我现在被封号的概率高吗?」*
[](LICENSE) [](https://claude.ai/code) [](https://python.org)
反封号.skill 是一个 **Claude Code 账号风险体检 skill**。 它会从网络出口、Claude API 路由、项目自动化、敏感数据、地区主体、审计记录几个方向, 给你的 Claude Code 使用环境做一次综合封号风险评估。
你不需要懂 IP2Location、ASN、Cloudflare colo、policy gate。 装好后直接问:
测试一下我目前账号被封的可能性
它会自己检查能检查的部分, 把不能确定的部分列成「需要你确认」。
[安装](#-安装) · [用法](#-用法) · [会检查什么](#-会检查什么) · [评分](#-评分) · [English](#english)
✨ 效果示例
你 ❯ 测试一下我目前 Claude Code 账号被封的可能性
反封号.skill ❯ 风险等级:watch
总风险分:52/100
已验证:
- api.anthropic.com 当前命中 NRT Cloudflare edge
- 检测到 split egress:不同 IP echo endpoint 返回了 2 个公网出口
- 当前项目存在 Telegram bridge + long-running agent
主要风险:
1. 网络出口不一致,Claude API 路径需要单独确认 egress/source IP
2. 长期 agent 自动化需要审计日志和外部动作确认机制
3. 地区/支付/手机号/主体信息无法从本地判断,需要用户确认
建议:
- 优先确认 api.anthropic.com 分流路径对应的公网出口
- 给外部动作加人工确认和日志
- 不要把 .env、token、敏感日志放进上下文
⚡ 安装
git clone https://github.com/swaylq/account-risk-skill.git \
~/.claude/skills/claude-code-account-risk
确认 Claude Code 能看到 skill 后,直接在 Claude Code 里问:
测试一下我目前账号被封的可能性
或:
用反封号 skill 检查这个项目
🚀 用法
你可以用自然语言触发:
测试一下我目前账号被封的可能性
帮我做一次 Claude Code 封号风险体检
这个项目跑 Claude Code 长期 agent 安全吗
我 Claude 账号被封了,帮我整理申诉材料
检查一下当前 IP 和 api.anthropic.com 路由
skill 会输出:
- 风险等级
- 总风险分
- 已验证项
- 需要用户确认项
- 高风险项
- 建议动作
- 如果已封号:申诉材料清单
🔍 会检查什么
自动检查
Claude Code 能自己检查:
- 当前项目是否有长期 agent、cron、daemon、Telegram bridge 等自动化入口
- 是否存在批量注册、批量消息、爬取、平台操纵、越权测试等高风险工作流
- 是否有
.env、token、API key、敏感日志、泄漏源码进入上下文的风险 - 是否有审计日志、pol
Related Skills
Dario
A local OpenAI- and Anthropic-compatible proxy that brings your Claude Pro/Max subscription to Cursor, Cline,
Screen Record
Record an MP4 screen video of a web flow end to end — intake (environment, URL, account, source) → confirm → p
Claude Grc Plugin
A Claude Code plugin that turns Claude into a senior GRC (Governance, Risk, and Compliance) analyst. 72+ refer
Checklist
Show compliance checklist for a risk tier
Strategy Bb Squeeze
Analyze and modify BB squeeze strategy — squeeze detection, breakout scoring, volume thresholds
Strategy Breakout
Analyze and modify breakout strategy — S/R detection, breakout confirmation, volume scoring
Related Agents
Audit Policy Compliance
Platform policy specialist. Returns schema-valid findings covering platform eligibility, regulated categories,
Deploy Preflight
Verifies a three.ws production deploy is safe to run BEFORE gcloud builds submit. Checks the load-bearing buil
Evasion
Antivirus and EDR evasion specialist for authorized red team engagements. Handles AMSI bypass, payload obfusca