agent-factory

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns detected across the 10 threat categories.\n- [Indirect Prompt Injection] (SAFE): The skill generates instructions for sub-agents that will ingest untrusted user data. While the templates do not include explicit boundary markers or sanitization logic, this is a known characteristic of the agentic framework it supports.\n
  • Ingestion points: User prompts to generated agents.\n
  • Boundary markers: Absent in templates.\n
  • Capability inventory: Bash, Write, Edit, Grep, Glob (defined in agent_generator.py).\n
  • Sanitization: Absent.\n- [Command Execution] (SAFE): The skill recommends 'Implementation' and 'Quality' agent types with access to the Bash tool. This is the intended primary purpose of the skill to enable functional automation and development tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:07 PM