wewe-rss-deploy

部署 WeWe RSS 服务，从微信公众号获取文章并生成 RSS，支持 SQLite 配置、微信读书登录和后台运行管理。

Security Scan

The instructions and required actions match the stated goal of deploying the WeWe RSS project; nothing in the skill metadata or SKILL.md appears to be trying to do something unrelated, though there are privacy and supply-chain considerations to review before running the steps.

Purpose & Capability

The SKILL.md precisely describes cloning the cooderl/wewe-rss GitHub repo, switching to a SQLite schema, installing Node dependencies, building, and running the service on port 4000. These steps are coherent with the skill name/description (deploying WeWe RSS). The skill metadata claims no required env vars; the runtime instructions do create a local .env for the service (AUTH_CODE, DATABASE_URL, etc.), which is expected for a self-hosted deployment.

Instruction Scope

Instructions stay within the deployment scope (git clone or GitHub zip download, pnpm install, prisma generate/migrate, build, start with pm2). Notable scope items to review: 1) the .env sets PLATFORM_URL to a third-party weread proxy (https://weread.111965.xyz) — using that public proxy may cause user login tokens or article content to be routed to an external service; 2) steps write runtime files under ~/.openclaw/workspace and a tools config file, and they instruct to remove/rename prisma directories inside the project (destructive only inside the cloned repo). Overall these actions are consistent with deployment but include privacy implications.

Install Mechanism

This is an instruction-only skill (no install spec) that instructs the user/agent to clone from GitHub and run pnpm/npx commands. The GitHub URLs used are normal project URLs (no shorteners or unknown hosts). The workflow will pull packages from npm during pnpm install, which is a normal supply-chain risk for any Node project but expected for this purpose.

Credentials

The skill does not request cloud credentials or unrelated secrets. It does instruct creating a local .env including AUTH_CODE and DATABASE_URL (local SQLite file). The main concern is the PLATFORM_URL pointing to a public third-party service that the project uses for WeRead forwarding — that service could receive or process authentication/login flow or article data. If you prefer to keep credentials/content private, you should self-host that proxy or avoid that PLATFORM_URL.

Persistence & Privilege

always:false (no forced global inclusion). The instructions recommend using PM2 to run the service and call pm2 save / pm2 startup, which will create a persistent startup entry (systemd / init modifications). That's a normal deployment behavior but has system-level persistence implications; installing pm2 globally may require elevated permissions on some systems. The skill writes a local path file under ~/.openclaw/workspace/tools for later discovery by other skills — this is limited to the user's home workspace and is within scope.

Assessment

This skill is a straightforward deployment recipe for the public wewe-rss project, but take these precautions before running it: 1) Review the upstream repository (https://github.com/cooderl/wewe-rss) yourself so you know what code will run. 2) Be cautious about PLATFORM_URL (https://weread.111965.xyz): it is a third-party proxy used for WeRead-related actions — it may see account/login actions or scraped content. If you care about privacy, self-host that proxy or remove/replace PLATFORM_URL. 3) The steps will fetch packages from npm and GitHub (normal supply-chain risk); consider running in an isolated VM/container. 4) Back up anything important before running destructive file commands; the script removes/renames prisma directories within the cloned project only, but confirm you are operating in the intended workspace path. 5) PM2 startup creates system-level persistence (systemd/init) and may require privileges — verify and review the startup unit created. If you want a safer test, run the service manually (pnpm start) in a container or non-privileged environment and inspect network activity and logs before enabling automatic startup.

SKILL.md - 部署 WeWe RSS 微信公众号 RSS 服务

触发条件

当 WeWe RSS 服务未运行（端口 4000 无响应），或用户要求部署 WeWe RSS 项目时激活。

功能概述

部署 WeWe RSS 服务，通过微信读书获取微信公众号文章，生成 RSS 订阅源。

前置条件

• Node.js >= 16
• pnpm 已安装
• Python 3.x

项目信息

• GitHub：https://github.com/cooderl/wewe-rss
• 安装后路径：~/.openclaw/workspace/wewe-rss-main

---

部署步骤

Step 1：克隆项目

# 克隆到 workspace
git clone https://github.com/cooderl/wewe-rss.git ~/.openclaw/workspace/wewe-rss-main

如果 Git 克隆失败，使用 curl 下载压缩包：

# 下载
curl -L "https://github.com/cooderl/wewe-rss/archive/refs/heads/main.zip" -o /tmp/wewe-rss.zip

# 解压（自动生成 wewe-rss-main 目录）
unzip -o /tmp/wewe-rss.zip -d ~/.openclaw/workspace/

# 清理
rm /tmp/wewe-rss.zip

Step 2：安装依赖

cd ~/.openclaw/workspace/wewe-rss-main
pnpm install

Step 3：配置 SQLite 数据库

切换 Prisma schema（项目默认是 MySQL，需要改为 SQLite）：

# 进入 server 目录
cd ~/.openclaw/workspace/wewe-rss-main/apps/server

# 检查当前 prisma 目录
ls prisma/

# 删除 MySQL schema（如果存在）
rm -rf prisma/

# 重命名 SQLite schema
mv prisma-sqlite prisma

创建数据目录：

mkdir -p ~/.openclaw/workspace/wewe-rss-main/apps/server/data
mkdir -p ~/.openclaw/workspace/wewe-rss-main/apps/data

Step 4：配置环境变量

创建配置文件 ~/.openclaw/workspace/wewe-rss-main/apps/server/.env：

HOST=0.0.0.0
PORT=4000

# 使用 SQLite
DATABASE_URL="file:../data/wewe-rss.db"
DATABASE_TYPE="sqlite"

# 访问授权码（自定义，建议设置复杂密码）
AUTH_CODE=your-secret-code-here

# 自动提取全文内容
FEED_MODE="fulltext"

# 服务地址
SERVER_ORIGIN_URL=http://localhost:4000

# 定时更新 Cron（每天 5:35 和 17:35 更新）
CRON_EXPRESSION="35 5,17 * * *"

# 微信读书转发服务（公共可用）
PLATFORM_URL="https://weread.111965.xyz"

# 关闭 HTML 清理（保留完整内容）
ENABLE_CLEAN_HTML=false

# 更新延迟（秒）
UPDATE_DELAY_TIME=60

# 每分钟最大请求数
MAX_REQUEST_PER_MINUTE=60

Step 5：初始化数据库

cd ~/.openclaw/workspace/wewe-rss-main/apps/server

# 设置环境变量
export DATABASE_URL="file:../data/wewe-rss.db"
export DATABASE_TYPE="sqlite"

# 生成 Prisma Client
npx prisma generate

# 初始化数据库
npx prisma migrate deploy

Step 6：构建项目

cd ~/.openclaw/workspace/wewe-rss-main
pnpm run -r build

如果构建时出现 MySQL 相关错误，重新检查 Step 3：

# 确认 prisma 目录是 sqlite 版本
cat ~/.openclaw/workspace/wewe-rss-main/apps/server/prisma/schema.prisma | grep provider
# 应该显示：provider = "sqlite"

Step 7：启动服务

直接运行（当前 session）：

cd ~/.openclaw/workspace/wewe-rss-main
pnpm run start:server

后台运行（PM2）：

# 安装 PM2（如果未安装）
npm install -g pm2

# 启动服务
cd ~/.openclaw/workspace/wewe-rss-main
pm2 start apps/server/dist/main.js --name wewe-rss

# 开机自启
pm2 save
pm2 startup

Step 8：保存安装路径

为了后续 Skill 能够找到项目，将路径写入配置：

mkdir -p ~/.openclaw/workspace/tools/
echo "~/.openclaw/workspace/wewe-rss-main" > ~/.openclaw/workspace/tools/wewe-rss-config.txt

Step 9：验证部署

检查服务是否运行：

# Windows
netstat -ano | findstr ":4000"

# Mac/Linux
lsof -i :4000

访问 Web UI：http://localhost:4000

---

使用流程

1. 打开 http://localhost:4000
2. 点击「账号管理」→「添加账号」→ 微信读书扫码登录
3. 进入「公众号源」→「添加」→ 粘贴公众号分享链接
4. 订阅后返回文章列表

---

API 接口

接口	说明
GET /feeds/{mp_id}.json?limit=10	获取公众号文章列表
GET /feeds/{mp_id}/{index}.json	获取指定文章正文
GET /api/v1/feeds	列出所有已订阅公众号

---

项目结构

~/.openclaw/workspace/wewe-rss-main/
├── apps/
│   ├── server/
│   │   ├── data/              # SQLite 数据库目录
│   │   │   └── wewe-rss.db
│   │   ├── dist/              # 构建输出
│   │   ├── prisma/            # 数据库 schema
│   │   └── .env               # 环境配置
│   └── data/                  # 数据文件软链接
└── package.json

---

常见问题

问题	解决方案
pnpm install 失败	检查 Node.js 版本，确保 >= 16
构建报错 schema 相关	确认已切换到 SQLite schema（Step 3）
数据库未创建	运行 npx prisma migrate deploy
服务无法访问	检查防火墙，确保 4000 端口开放
微信读书登录失败	使用推荐 PLATFORM_URL，或检查网络
重启后服务消失	使用 PM2 后台运行（Step 7）

---

卸载

# 停止服务
pm2 delete wewe-rss

# 删除项目
rm -rf ~/.openclaw/workspace/wewe-rss-main

# 清理配置
rm -f ~/.openclaw/workspace/tools/wewe-rss-config.txt

---

娆㈣繋鍏虫敞浣滆€呭井淇″叕浼楀彿锛?寮€璁板仛浜у搧*