blog-post

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to clone and operate on a public GitHub repo (git clone https://github.com/jim60105/blog.git) and to read repository files such as .github/instructions/quill-sage.instructions.md and references/writing-guidelines.md, which are user-authored third-party content that the agent must read and use to drive writing and publishing decisions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs cloning and using the remote repository https://github.com/jim60105/blog.git (including initializing submodules and running ./switch-site.sh inside the repo), so content fetched at runtime from that URL can execute code or directly control the agent's workflow.