close
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external CRM data (such as contacts and leads) which could potentially contain malicious content intended to influence agent behavior. * Ingestion points: Data retrieved from Close CRM endpoints via the Apideck API. * Boundary markers: Not explicitly defined within the connector skill; the skill references external best practices for architecture and handling. * Capability inventory: Network access to unify.apideck.com and the Close CRM API. * Sanitization: The provided code examples do not include explicit sanitization of the retrieved data.
- [SAFE]: The skill adheres to security standards by using environment variables for sensitive credentials (APIDECK_API_KEY, APIDECK_APP_ID). All external references and network operations target verified vendor infrastructure (apideck.com) or the official Close CRM documentation and API services.
Audit Metadata