b2c-bm-users-roles

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for executing administrative commands via the b2c (Salesforce B2C CLI) utility to manage instance-level resources such as users, roles, and access permissions.
  • [EXTERNAL_DOWNLOADS]: The skill references the @salesforce/b2c-cli package, which is an official tool provided by a recognized vendor (Salesforce). It suggests using npx to run the tool if it is not installed locally.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes and manages external data retrieved from the Salesforce OCAPI Data API, creating a surface for indirect instructions if the source data is compromised.
  • Ingestion points: Data is ingested from external API endpoints (e.g., /users, /roles) and local configuration files (e.g., dw.json, admin-perms.json) used for permission management.
  • Boundary markers: The instructions do not define specific delimiters for separating untrusted data from the agent's core instructions.
  • Capability inventory: The skill allows the agent to perform high-privilege operations including creating users, assigning the 'Administrator' role, and rotating authentication access keys.
  • Sanitization: The skill relies on the underlying b2c-cli and the Salesforce OCAPI Data API to perform validation and sanitization of the managed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:00 AM
Security Audit — agent-trust-hub — b2c-bm-users-roles