critique-brand-consistency

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The instructions are focused on brand critiquing and do not attempt to override agent safety guidelines or system prompts.
  • [DATA_EXFILTRATION]: The skill reads project-specific configuration files (mood.md, voice.md, tokens.md) for analysis. It lacks any network tools or commands to transmit data externally, and no hardcoded credentials or sensitive system paths are accessed.
  • [COMMAND_EXECUTION]: No shell commands, subprocesses, or privilege escalation attempts were detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by reading external files.
  • Ingestion points: Reads mood.md, voice.md, and tokens.md from the project directory.
  • Boundary markers: The skill does not define specific delimiters for the content of these files.
  • Capability inventory: The skill is limited to text analysis and critiquing; it has no access to tools for file writing, network communication, or code execution.
  • Sanitization: No input validation or sanitization is performed on the reference files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 05:32 PM
Security Audit — agent-trust-hub — critique-brand-consistency