spec-setup

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses a template file located at ~/.maestro/templates/project-tech.json. This access retrieves a configuration template from the user's home directory to facilitate project initialization.
  • [PROMPT_INJECTION]: The skill scans untrusted files in the local project directory to generate specification documents, creating a surface for indirect prompt injection.
  • Ingestion points: Scans repository files such as package.json, Cargo.toml, and .eslintrc.
  • Boundary markers: Not specified for the output files generated in .workflow/specs/.
  • Capability inventory: Employs Read, Write, Glob, Grep, and Bash for data processing and file creation.
  • Sanitization: No explicit sanitization or validation of the scanned file content is described in the implementation steps.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to perform environment checks, specifically verifying the existence of the .workflow directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 01:12 AM