erb-to-view-model

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing data from various local files, including ERB templates and external rule files located in @.cursor/rules/. Instructions embedded in these files could potentially manipulate the agent's logic during the refactoring process.
  • Ingestion points: The skill reads files from the app/ directory and local rule directories (@.cursor/rules/, @prompts/).
  • Boundary markers: No explicit boundary markers or instructions to disregard embedded commands are utilized when reading external files.
  • Capability inventory: The skill performs file system read/write operations and executes local shell commands (e.g., rspec).
  • Sanitization: The skill lacks mechanisms to sanitize or validate the content of ingested files before processing them.
  • [COMMAND_EXECUTION]: To verify the migration, the skill instructs the agent to execute rspec commands. This allows for the execution of code contained within the project's test suite and the newly generated ViewModels as part of the intended development workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 12:38 AM