Social Media Auto

自媒体内容生成技能 - 热点抓取 + AI 创作 + 多平台适配

What to consider before installing: - Expect this skill to contact external Alibaba Cloud services (mcporter / 阿里云 MCP) for both trending search and AI generation. The package metadata does not declare any credentials — you should only install if you understand and trust how you will provide mcporter/Aliyun credentials (and preferably configure them in a controlled environment). - The scripts call `npx mcporter ...` at runtime. `npx` can fetch packages on first run (supply-chain risk). If you don't want dynamic downloads, review and preinstall required tooling and pin versions. - The code uses child_process.execSync with string interpolation of user input (e.g. the --topic value). That is a command-injection hazard if you or other automated systems pass untrusted input into these scripts. Avoid running these scripts with untrusted inputs, or sanitize/escape arguments before use. - The SKILL.md included a prompt-injection-like artifact (unicode-control-chars). Inspect SKILL.md and the included files for any hidden characters or instructions that could alter behavior. - If you decide to try it: run it in a sandboxed environment (container or VM) first, inspect/verify mcporter configuration, and avoid running the test script or fetch/generate on any system containing sensitive credentials. If you need this functionality but want lower risk, ask the author to: declare required env vars, document mcporter setup, avoid shell interpolation (use child_process with argument arrays or a proper SDK), and pin any external dependencies.

Type: OpenClaw Skill Name: social-media-auto Version: 1.0.0 The skill bundle contains a shell injection vulnerability in 'scripts/generate-content.js' and 'scripts/fetch-trending.js'. User-provided input from the '--topic' argument is directly interpolated into a shell command executed via 'execSync' to call an external MCP tool (npx mcporter). While the overall logic for social media content generation appears legitimate and aligned with the stated purpose, the lack of input sanitization allows for arbitrary command execution. No evidence of intentional malice, data exfiltration, or persistence was found.