accessibility-general

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

• Third-party content exposure detected (high risk: 0.70). The skill's "AI Scraping Policy" in SKILL.md explicitly instructs the agent to fetch content "from any URL" (after consulting examples/TRUSTED_SOURCES.yaml), which means the agent may ingest open/public web pages and blogs (untrusted third‑party content) that can influence its decisions and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs falling back to the mgifford/ACCESSIBILITY.md examples directory (https://github.com/mgifford/ACCESSIBILITY.md), which the agent would fetch at runtime and whose example skill files directly control agent behavior/instructions, so this remote repository is a runtime dependency that can change prompts.