tech-stack-evaluator

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's primary function is analytical, operating on user-provided technology data. It does not perform any external network requests, fetch remote code, or access sensitive system files.
  • [COMMAND_EXECUTION]: The skill provides several Python scripts (e.g., stack_comparator.py, tco_calculator.py) for processing technology metrics. These scripts perform deterministic calculations and do not use unsafe functions like eval() or exec() on user input.
  • [PROMPT_INJECTION]: The skill includes an automated format detector (scripts/format_detector.py) that parses natural language and structured data. While this presents a surface for indirect prompt injection, the risk is assessed as safe because the parsed output is used for numerical scoring and structured report generation rather than instruction following.
  • Ingestion points: FormatDetector in scripts/format_detector.py processes raw user text, JSON, and YAML.
  • Boundary markers: None implemented in the processing scripts.
  • Capability inventory: ReportGenerator.export_to_file in scripts/report_generator.py allows writing reports to disk.
  • Sanitization: The detector normalizes keywords and extracts specific entities, effectively filtering non-relevant content from the analytical pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 01:08 AM