deel
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides the agent with the ability to ingest data from an external HRIS provider (Deel).
- Ingestion points: Data is retrieved via
apideck.hris.employees.listand related methods inSKILL.md. - Boundary markers: No specific delimiters or warnings for the agent to ignore instructions within the retrieved HRIS data are provided in the documentation.
- Capability inventory: The skill facilitates listing and reading employee, department, and time-off data, as well as arbitrary direct calls via the Proxy API.
- Sanitization: No explicit sanitization of the external API responses is mentioned in the implementation examples.
- [SAFE]: The skill follows secure development practices by recommending the use of environment variables (
APIDECK_API_KEY,APIDECK_APP_ID) for authentication instead of hardcoding credentials. All external references target official domains (apideck.com, deel.com) or trusted package registries.
Audit Metadata