cruise-control

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local script node scripts/run-ast-analysis.mjs analyze . during the Gear 1 phase. While this is a functional part of the developer tool, it represents a command execution pattern on the local environment.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests and processes untrusted data from the user's project to drive subsequent automated actions.
  • Ingestion points: The skill reads all project files for analysis in Gear 1 and documentation extraction in Gear 2.
  • Boundary markers: There are no explicit boundary markers or instructions mentioned to prevent the agent from following directions embedded within the source code being analyzed.
  • Capability inventory: The skill can execute local scripts (node scripts/run-ast-analysis.mjs), read the filesystem, and write new code/specifications via the /speckit.implement and /speckit.tasks commands.
  • Sanitization: No sanitization or filtering of the analyzed source code is described before the content is passed to downstream LLM agents for specification and implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 08:14 AM