figma-skill

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). The URL points to an unverified, personal GitHub repository offered as an npx-installable "experimental" skill—not an official vendor or well-known project—so it could execute arbitrary code at install/run time and should be treated as potentially unsafe without review.

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow “connects to the Figma API to discover Figma Community file resources” and then “extracts design tokens from those files,” which means it ingests outsider-authored public/community Figma content (public web/community sources) into the agent’s LLM context via the extracted token text.

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

• Potentially malicious external URL detected (high risk: 0.90). The listed install step ("npx skills add nafiurrahmanniloy/figma-skill") fetches and executes a remote package tied to the upstream repository (https://github.com/nafiurrahmanniloy/figma-skill), so runtime installation requires executing external code from that URL.