aic-journey-design
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for designing journeys that ingest untrusted user input via Platform Username and Attribute Collector nodes (e.g., in social-and-local-registration-authentication.md). While it specifies boundary markers like anti-enumeration wiring and validation policies, the presence of capabilities such as Scripted Decision nodes (server-side JS execution) and Patch Object nodes (identity store writes) across the reference files creates a surface for indirect prompt injection. Sanitization is addressed through the use of built-in validation nodes and environment-specific variables.
- [DYNAMIC_EXECUTION]: The skill facilitates the execution of server-side JavaScript or Groovy logic via the Scripted Decision node. It provides guidance on state management and logic branching while recommending the use of Environment-Specific Variables (ESVs) to protect sensitive configuration and maintain portability across environments.
- [COMMAND_EXECUTION]: The skill leverages platform-native MCP tools, such as createJourney and updateJourneyNode, to manage journey configurations. These operations are restricted to the connected PingOne Advanced Identity Cloud environment and are consistent with the skill's primary purpose of identity journey design.
Audit Metadata