experience-ui-bundle-custom-app-generate

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Salesforce CLI (sf) to query organization namespaces and validate metadata deployments, which is the intended functionality for this skill type.
  • [COMMAND_EXECUTION]: The skill executes a local script scripts/resolve-uibundle-path.mjs via Node.js to find the location of existing configuration files.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the local file system (directory names) and user input to generate XML metadata.
  • Ingestion points: Directory names in uiBundles/ and user-provided appLabel strings in SKILL.md.
  • Boundary markers: Absent from the XML templates.
  • Capability inventory: File system writes to applications/ and execution of sf CLI tools.
  • Sanitization: The resolve-uibundle-path.mjs script implements a strict alphanumeric regex (/^[A-Za-z0-9][A-Za-z0-9_-]*$/) to sanitize the appName variable before it is used in file system operations, effectively mitigating path traversal risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 11:53 PM