sync-global-rules

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill directly fetches and installs files from the public GitHub repository nangongwentian-fe/Awesome-GlobalRule (see scripts/check_updates.sh and scripts/sync.sh which call gh api to download AGENTS.md and docs/agent-rules/*) and writes them into local agent rule directories, so untrusted third‑party content can materially change agent behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). At runtime the scripts use gh api to fetch files from the GitHub repository nangongwentian-fe/Awesome-GlobalRule (e.g. https://api.github.com/repos/nangongwentian-fe/Awesome-GlobalRule/contents/AGENTS.md or https://github.com/nangongwentian-fe/Awesome-GlobalRule) and then overwrite local agent rule files that directly control AI agent prompts/instructions, so the remote content directly controls agent behavior and is a required dependency.