zendesk
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an ingestion surface where the agent processes raw response data from an external API, which could theoretically contain instructions. * Ingestion points: Raw response data from the Apideck Proxy endpoint at unify.apideck.com (SKILL.md). * Boundary markers: Absent; no specific delimiters are defined to separate ingested data from agent instructions. * Capability inventory: Network operations via HTTP requests (SKILL.md). * Sanitization: Absent; no explicit validation or escaping of the external API response content is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to make network requests to unify.apideck.com and provides references to official documentation at zendesk.com and apideck.com. These are verified official domains for the services and vendors described in the skill.
Audit Metadata