NYC

project-planner

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: CRITICALPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill is designed to generate requirements, design, and implementation plans that serve as blueprints for AI-assisted development, creating a vector where malicious data in the generated docs could influence downstream AI tasks.
  • Ingestion points: User-defined project names, features, and component descriptions in requirements-template.md and scripts.
  • Boundary markers: The template uses bracketed placeholders (e.g., [PROJECT NAME]) but lacks specific delimiter instructions to prevent the agent from obeying embedded instructions.
  • Capability inventory: The skill generates task lists and implementation plans used to drive autonomous coding agents.
  • Sanitization: Not verifiable as the generation scripts (generate_project_docs.py) were not provided.
  • [NO_CODE] (SAFE): Missing Functional Scripts. The skill refers to 'scripts/generate_project_docs.py' and 'scripts/validate_documents.py' as the primary execution method, but these files are not included in the skill folder.
  • [EXTERNAL_DOWNLOADS] (SAFE): No network operations or external downloads are present in the provided files. The scanner alert for requirements.md indicates a potential threat in the generated output or repository that could not be confirmed in the provided assets.
Recommendations
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 17, 2026, 06:09 PM