impeccable

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs agents to load and use public third-party content — e.g., it tells agents to source stock images from Unsplash via public URLs (reference/brand.md) and to navigate to arbitrary web pages and inject/read overlays/console output during the live critique flow (reference/critique.md), which exposes the agent to untrusted, user-provided web content that can influence actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The critique flow invokes npx (e.g., "npx impeccable --json" / "npx impeccable live &"), which at runtime fetches and executes the "impeccable" package from the npm registry (e.g. https://registry.npmjs.org/impeccable), so remote code is fetched-and-run and required for the detector/live steps.