project-docs-setup

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • NO_CODE (SAFE): The skill files are exclusively documentation templates. There are no shell scripts, Python/Node.js code, or binary executables within the skill package. This significantly reduces the attack surface for most threat categories.
  • PROMPT_INJECTION (SAFE): While the templates contain instructional language for an AI agent (e.g., "Mandatory requirement: all code development must follow TDD"), these are legitimate task-specific constraints and do not attempt to bypass LLM safety filters or disregard system instructions.
  • DATA_EXFILTRATION (SAFE): No network operations (curl, wget, etc.) or access to sensitive file paths were found. The templates use placeholders for documentation but do not hardcode secrets.
  • INDIRECT_PROMPT_INJECTION (SAFE): Although the templates define placeholders (e.g., {USER_STORIES}, {PROJECT_NAME}) which are interpolation surfaces for external data, the skill does not include any logic for data ingestion or command execution. The surface is strictly for static documentation generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:05 PM