tech-stack-evaluator

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input which can contain instructions aimed at manipulating agent behavior.
  • Ingestion points: The FormatDetector class in scripts/format_detector.py parses user-provided input from several formats including natural language text, YAML, and JSON.
  • Boundary markers: The implementation lacks explicit delimiters or specific instructions to the agent to treat the processed data as untrusted content.
  • Capability inventory: The skill includes a file-writing capability through the ReportGenerator.export_to_file method in scripts/report_generator.py.
  • Sanitization: There is no evidence of string sanitization or content validation performed on the user-supplied data before it is utilized in analysis or written to files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 02:34 PM
Security Audit — agent-trust-hub — tech-stack-evaluator