launch-signals

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

• Third-party content exposure detected (low risk: 0.10). The required workflow searches public regulatory registers and certificate logs, which contain outsider-authored text such as trademark goods/services descriptions and device filings.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs installation of an MCP plugin from an arbitrary third-party GitHub repository (un-seen/hedwigai-skills) and connection to a remote MCP endpoint (mcp.hedwigai.com), introducing weak/unknown operational dependencies.