form-cro

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a local Python utility (scripts/form_field_analyzer.py) to perform static analysis on form code. This script is bundled with the skill and operates solely on locally provided files without executing the content of those files.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests data from external context and data files.
  • Ingestion points: The skill reads marketing context from .claude/product-marketing-context.md and form structure data from user-provided files (processed via form_field_analyzer.py).
  • Boundary markers: The instructions do not implement specific delimiters or safety instructions to prevent the agent from following directives that might be embedded in the analyzed forms or marketing context.
  • Capability inventory: The skill possesses the capability to execute internal scripts and read files within the workspace.
  • Sanitization: No specific filtering or sanitization logic is applied to the content of the processed files before it enters the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:17 PM
Security Audit — agent-trust-hub — form-cro