jep-jeanne-sulzer
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, obfuscation, or unauthorized network operations were detected. The skill's operations are strictly defined for legal research and auditing of documents related to Colombian transitional justice. The network resources referenced (jep.gov.co, unmissions.org) are official government and international organization domains, which are recognized as safe and authoritative sources.- [PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection as it is designed to ingest user-supplied legal documents for auditing and fetch content from external authoritative websites. While this functionality is necessary for its purpose, it relies on the model's internal safety guardrails to mitigate any malicious instructions embedded in processed texts.
- Ingestion points: User-provided legal documents (audit mode) and data retrieved from official repositories via web_fetch.
- Boundary markers: The skill does not implement specific delimiters to isolate external content from instructions.
- Capability inventory: The agent is instructed to use web_fetch to retrieve information from specified domains.
- Sanitization: No explicit sanitization or filtering is described; accuracy is managed through a manual verification-first workflow.
Audit Metadata