broll-master-build

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data (SRT and design files) as part of its core workflow, creating a potential surface for indirect prompt injection. Ingestion points: SRT files, design files, and shared asset indices. Boundary markers: Conceptual 'truth' is defined as immutable. Capability inventory: File writing (HTML) and command execution. Sanitization: None mentioned.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute an assignment-provided 'standard command', confirming command-line capabilities within the environment.
  • [SAFE]: Instructions prohibiting the creation of inspection, trace, or proof tools are interpreted as workflow constraints to prevent redundant artifact generation rather than malicious obfuscation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 07:43 AM
Security Audit — agent-trust-hub — broll-master-build