cascade-hr
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or metadata.
- [CREDENTIALS_UNSAFE]: The skill correctly instructs the use of environment variables (APIDECK_API_KEY, APIDECK_APP_ID) for credential management. No hardcoded secrets or sensitive configuration files are accessed.
- [PROMPT_INJECTION]: The skill defines a standard data ingestion surface for processing external HRIS records. While this introduces a potential for indirect prompt injection, it is the primary purpose of the skill and no exploitable capability chains were found.
- Ingestion points: API responses from IRIS Cascade HR (SKILL.md)
- Boundary markers: Absent
- Capability inventory: Network requests to Apideck's unified API
- Sanitization: Not explicitly specified
- [EXTERNAL_DOWNLOADS]: The skill references the '@apideck/unify' Node.js package and fetches OpenAPI specifications from official vendor domains (specs.apideck.com). These resources are associated with the vendor's own infrastructure.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns, dynamic execution, or unauthorized script downloads were identified.
Audit Metadata