apideck-unified-api
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an integration model that processes data from over 146 external SaaS providers. This introduces a surface for indirect prompt injection where data from external APIs could potentially contain instructions for the agent.
- Ingestion points: External data retrieved from third-party SaaS connectors via the Unified API (SKILL.md).
- Boundary markers: The skill does not explicitly define prompt delimiters for handling untrusted data.
- Capability inventory: The integration relies on SDKs and tools that execute network requests and process structured data (SKILL.md).
- Sanitization: Standard documentation does not specify sanitization routines for the ingested data.
Audit Metadata