Back to plugin

Security audit

BrandClaw: Forensic Voice Enforcer

Security checks for vulnerabilities and agentic risk

Overview

BrandClaw is a disclosed local brand-voice hook that changes agent style and prompts, with no evidence of network exfiltration or credential access.

Use BrandClaw if you want an agent-wide brand voice filter. Be aware that it can override normal assistant tone, the advertised brand/strictness options may not all be implemented, and the external demo link is separate from the local plugin behavior.

Static analysis

No suspicious patterns detected.