general-video

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill tells the agent to run npx hyperframes auth status and "relay its output verbatim" (SKILL.md:13), which could cause the model to print sensitive tokens or credentials returned by that command; it also tells the agent to "paste them in full" when dispatching worker packets (SKILL.md:94), which can likewise expose secrets embedded in files or packets.

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text can enter at runtime via the user/tenant-provided STORYBOARD.md frame blocks (which are inlined into the generated scene packet consumed by the delegated frame worker), so a user can post poison content into the project files without any prior selection step.