content-translator

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). This is a personal GitHub repository (peter-tu-zynkr) referenced for installation with npx, which would fetch and execute code from an unverified source—GitHub repos from unknown accounts with no community signals are potentially risky.

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 is a translation service that will ingest user-provided source text (and possibly any fetched/forwarded content the user supplies) into the LLM context; that source text is outsider-authored free-form prose/comments, creating indirect prompt-injection exposure.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's install command uses npx to fetch and run remote code from https://github.com/peter-tu-zynkr/zynkr-skill-builder (npx skills add ...), which downloads and executes external code that defines the skill's prompts/behavior, so this is a runtime external dependency.