sync-local-global-agents-skills

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

• Third-party content exposure detected (medium risk: 0.30). 脚本在运行时会先通过 src/memorix.ts 的 fetchFromGitHub/fetchTree 从 GitHub（Content API 与 raw 下载）读取 memorix skills 的任意文件内容并写入到 ~/.agents/skills，随后这些内容被同步到各本地平台目录，因此存在将“外部（GitHub 仓库）作者的 free text/文件内容”间接喂给后续流程的风险路径。

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

• Potentially malicious external URL detected (high risk: 0.90). The code calls the GitHub Content API at runtime (https://api.github.com/repos/AVIDS2/memorix/contents/plugins/${agent}/memorix/skills?ref=${ref}) and downloads raw files (via raw.githubusercontent.com) which are written into ~/.agents/skills and can contain agent skill prompts/instructions that control agent behavior.