zendesk-sell
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external CRM records (contacts, leads, opportunities, etc.) which represents an attack surface for indirect prompt injection.
- Ingestion points: Data enters the agent context through methods like
apideck.crm.contacts.list,leads.list, and other CRM search/read operations described inSKILL.md. - Boundary markers: The provided code examples do not explicitly demonstrate the use of delimiters or 'ignore embedded instructions' warnings when processing CRM field values.
- Capability inventory: The skill allows the agent to read, write, and search across contacts, companies, leads, opportunities, activities, and pipelines in Zendesk Sell.
- Sanitization: No specific sanitization or filtering logic is shown in the minimal TypeScript example, meaning instructions embedded in CRM fields (e.g., a contact's 'Note' field) could potentially influence the agent.
Audit Metadata