baoyu-multimedia-suite

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The repository includes explicit, deliberate techniques to bypass web platforms' anti-automation defenses and to drive actions using persistent logged-in browser profiles and system-level clipboard/keystroke injection, which enable automated posting under authenticated sessions and can be abused for unauthorized or covert account activity.

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime path “/baoyu-comic posts/turing-story/source.md” indicates the agent will read user-provided free text from a local markdown/source file (e.g., path/to/content.md/source.md) and feed it into the comic generation workflow, so outsiders can submit arbitrary text that the LLM ingests.