update-agent-skills

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

• Third-party content exposure detected (high risk: 0.90). The fallback flow explicitly instructs reinstalling skills from public repositories (e.g., the example npx skills add https://github.com/github/awesome-copilot --skill git-commit), which fetches and installs untrusted third-party code/content that the agent will ingest and which can materially change its behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The fallback flow explicitly runs a runtime reinstall from a remote git repository (example: https://github.com/github/awesome-copilot via npx skills add), which fetches and installs remote skill code that can directly control agent prompts or execute code.