init-translator

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

• Third-party content exposure detected (high risk: 0.85). The required workflow instructs fetching and ingesting two public Zotero “translator” pages at runtime (public web content), which can include free-form text that the agent reads into its LLM context via the fetch/ingest step.

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

• Potentially malicious external URL detected (high risk: 1.00). This skill explicitly instructs the agent to fetch and ingest the remote documents at https://www.zotero.org/support/_export/raw/dev/translators and https://www.zotero.org/support/_export/raw/dev/translators/coding at runtime, which directly supplies content that would control prompts/instructions and is required for the skill to operate.