implement-to-pixel

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests user-provided Confluence PRD links (*.atlassian.net/wiki/...) and Figma URLs (figma.com) — via the mekari-taste and figma-implement-design flows described in SKILL.md — and the agent is required to read and interpret that content to decide patterns, states, and generate code, so untrusted third‑party content could materially influence its actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches user-provided Confluence pages (*.atlassian.net/wiki/...) and Figma files (figma.com/file/... or figma.com/design/...) at runtime and injects that content into the model's imagination/generation pipeline, so external content directly controls agent instructions and is a required dependency.