adp-ihcm
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted HRIS data (employees, payroll) from ADP iHCM. Malicious instructions within these records could influence agent behavior.\n
- Ingestion points: HRIS data retrieval methods such as employees.list, as described in SKILL.md.\n
- Boundary markers: The skill does not define specific delimiters or instructional guards to prevent the agent from executing instructions found within the data.\n
- Capability inventory: Data reading and synchronization capabilities are demonstrated, which could be chained with other agent tools.\n
- Sanitization: There is no evidence of data validation or sanitization before the information is processed by the agent.\n- [SAFE]: All network operations and documentation links refer to official, verified domains belonging to Apideck and ADP.\n- [SAFE]: Credential management in code examples follows the best practice of using environment variables rather than hardcoded secrets.
Audit Metadata