generating-sorbet-inline

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill analyzes local Ruby source files to infer type information, which creates an attack surface for indirect prompt injection via malicious code comments or documentation blocks.\n
  • Ingestion points: Processes Ruby source files during the analysis phase (Step 1).\n
  • Boundary markers: No explicit instructions are provided to the agent to treat input file contents as untrusted data or to disregard instructions found within them.\n
  • Capability inventory: Executes shell commands ('srb tc' and 'rm'), reads project files, and performs write operations to modify Ruby source code.\n
  • Sanitization: There is no defined process for sanitizing or validating the contents of the Ruby files before analysis.\n- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands to manage its state and run validation tools.\n
  • Evidence: Step 5 instructs the execution of 'srb tc' or 'bundle exec srb tc'.\n
  • Evidence: The 'Multi-File Processing' section requires the agent to use 'rm .sorbet-inline-generation-todo.tmp' for cleanup.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 12:31 AM