skills/tonone-ai/tonone/brace-kb/Gen Agent Trust Hub

brace-kb

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes Bash commands to identify relevant data sources for its audit.
  • Evidence: find . -name "*.md" -o -name "*.csv" -o -name "*.txt" 2>/dev/null | xargs grep -l "ticket\|question\|issue\|bug\|report" in SKILL.md.
  • Evidence: find . -name "faq*" -o -name "help*" -o -name "kb*" -o -name "docs*" in SKILL.md.
  • These commands are limited to read-only file discovery operations and do not pose a risk to system integrity.
  • [DATA_EXFILTRATION]: Although the skill identifies files that could contain customer support data, there are no instructions or scripts designed to exfiltrate this information to an external endpoint.
  • [PROMPT_INJECTION]: No patterns of behavior override or safety bypass were detected. The persona 'Brace' is used within the context of the support engineer role and does not attempt to subvert the agent's core instructions.
  • [SAFE]: The skill's operations are transparent and align perfectly with its stated goal of knowledge base management. It follows best practices by requesting confirmation from the user ('validate with the founder') when data is missing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 02:48 PM
Security Audit — agent-trust-hub — brace-kb