5w1h-decision

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to execute local Python scripts (generate_token.py and validate_5w1h.py) using the uv tool to manage session tokens and validate framework compliance.
  • [EXTERNAL_DOWNLOADS]: The installation instructions utilize curl to fetch the skill's markdown definition from a public GitHub repository.
  • [PROMPT_INJECTION]: The skill defines strict behavioral constraints through 'Avoidance Language Detection,' blocking specific phrases (e.g., 'too complex', 'for now') to prevent the agent from bypassing established quality standards.
  • [PROMPT_INJECTION]: Indirect Prompt Injection vulnerability surface exists as the skill is intended to process external requirement documents and user-provided scenarios.
  • Ingestion points: Data enters the system via the Why section, which references external files like docs/app-requirements-spec.md.
  • Boundary markers: The skill uses a structured Markdown template to organize fields, but lacks explicit safety delimiters to prevent the agent from following instructions embedded within the processed requirement files.
  • Capability inventory: The skill can execute local Python scripts and generate file content for decision tokens.
  • Sanitization: Validation is performed via an internal script (validate_5w1h.py) and a keyword-based blocklist targeting avoidance language.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 06:39 PM