navan-enterprise-rbac

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill provides TypeScript and Bash script templates designed to be executed by the agent to interact with the Navan API.
  • [DATA_EXFILTRATION]: Contains network operations targeting api.navan.com. This is a well-known service and the interactions are consistent with the skill's administrative purpose.
  • [INDIRECT_PROMPT_INJECTION]: Processes user-provided email lists during bulk role assignment, which is a potential surface for indirect injection. 1. Ingestion points: userEmails parameter in assignDepartmentRoles function (SKILL.md). 2. Boundary markers: None present. 3. Capability inventory: Performs network requests using fetch (SKILL.md). 4. Sanitization: No input validation or sanitization observed for email strings.
  • [COMMAND_EXECUTION]: Includes curl commands for auditing and reporting, which interact with the system shell and the Navan API.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 12:37 PM
Security Audit — agent-trust-hub — navan-enterprise-rbac