form-edit

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local PowerShell script (form-edit.ps1) to perform XML processing. This is the primary intended function of the skill and the execution is restricted to the local environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes JSON data which acts as a data ingestion surface for modifying XML files.\n
  • Ingestion points: The JsonPath parameter in both PowerShell and Python scripts.\n
  • Boundary markers: None; the tool expects the JSON to be provided as a local file.\n
  • Capability inventory: File system write access via [System.IO.File]::WriteAllText and Python's open() to update the Form.xml file.\n
  • Sanitization: Both implementations use XML escaping for user-provided text content (e.g., titles and hints).\n- [REMOTE_CODE_EXECUTION]: The Python implementation imports the lxml package for XML parsing. No remote code execution or untrusted downloads were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 06:01 PM
Security Audit — agent-trust-hub — form-edit