video-podcast-maker

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

• Third-party content exposure detected (high risk: 0.75). Step 2 “Research Topic” explicitly uses the agent’s web search/fetch capabilities and saves fetched content into videos/{name}/topic_research.md, which is then available for the agent to read and include in its LLM context (public web content fetched at runtime).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's auto-update logic runs git fetch/pull against the skill's repository (origin) — e.g. https://github.com/Agents365-ai/video-podcast-maker — during runtime and (with user consent) can pull remote code that would alter the skill's scripts and behavior, meaning a remote git URL is used at runtime to fetch content that can change/execute agent instructions.