ui-setup

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The skill includes explicit instructions that agents "re-read" and "must obey" the generated STYLESEED.md (and "you'll obey it on every prompt"), which attempts to impose persistent, out-of-scope behavior on the agent and thus is a prompt injection.

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

• Third-party content exposure detected (low risk: 0.10). Step 3 conditionally fetches and reads https://raw.githubusercontent.com/VoltAgent/awesome-design-md/main/design-md/[brand]/DESIGN.md whose contents are free text from a third-party (GitHub) author, and the agent ingests/extracts it at runtime without further restricting the text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs at runtime to fetch and parse remote content from https://raw.githubusercontent.com/VoltAgent/awesome-design-md/main/design-md/[brand]/DESIGN.md to extract colors that the agent then applies to project files, so the external URL directly controls the agent's generated output.