nexthink-campaign-translator
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to read and translate text content from external Nexthink campaign export files.
- Ingestion points: The agent processes .json files provided by the user, specifically reading campaign question text and choice labels (SKILL.md, Step 1).
- Boundary markers: While the skill utilizes JSON structure to isolate data fields, it lacks explicit delimiters or instructions to the model to treat the ingested text as untrusted data when performing translations.
- Capability inventory: The skill possesses the ability to read from and write to the local file system using the add_translations.py script. It does not have network access or the ability to execute arbitrary shell commands.
- Sanitization: The companion script add_translations.py includes robust validation for HTML structure, preservation of link URLs (href), and protection of template tokens (e.g., {{user}}). However, it does not implement any mechanisms to detect or sanitize malicious instructions that may be embedded within the translatable text nodes.
Audit Metadata