optise-helix-eu-trust-centre
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes an 'Audit existing' mode (Step 0) which processes content from user-provided URLs using
web_fetch. This creates an indirect prompt injection surface where instructions embedded in external web pages could attempt to influence the agent's behavior. - Ingestion points: External URLs processed during the audit workflow.
- Boundary markers: None explicitly defined in the instructions for the fetched content.
- Capability inventory: The skill utilizes
web_fetchandweb_searchfor verification and auditing. - Sanitization: No explicit sanitization or escaping of external content is described beyond the instruction to 'flag gaps only' in audit mode.
- [SAFE]: The skill explicitly forbids the invention of sensitive data such as data residency, subprocessors, or security certifications, mandating the use of
[User to add: ...]placeholders. This follows security best practices for handling compliance and legal information.
Audit Metadata