init-translator
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The skill instructs the agent to fetch documentation from zotero.org and target webpages, which involves network requests to non-whitelisted domains.\n- [COMMAND_EXECUTION] (LOW): The agent is explicitly directed to 'use curl in the shell' for network activity, granting command-line access for data retrieval.\n- [PROMPT_INJECTION] (LOW): Ingesting external documentation and arbitrary target webpages creates a surface for Indirect Prompt Injection. (1) Ingestion points: Remote Zotero documentation and webpages fetched via curl (SKILL.md). (2) Boundary markers: Absent. (3) Capability inventory: Shell access (curl) and local file read/write (translator scripts). (4) Sanitization: Absent.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata