brand

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/sync-brand-to-tokens.cjs utilizes child_process.execSync to execute a local Node.js script for design token generation. The execution path is hardcoded and points to a relative location within the project directory.\n- [PROMPT_INJECTION]: The scripts/inject-brand-context.cjs script aggregates brand guidelines from docs/brand-guidelines.md into a structured prompt block for the agent. This creates a surface for indirect prompt injection.\n
  • Ingestion points: Content is dynamically read from docs/brand-guidelines.md during execution.\n
  • Boundary markers: The script wraps extracted data with explicit section headers such as BRAND CONTEXT: and BRAND VOICE:.\n
  • Capability inventory: The skill has permissions for file system operations and execution of specific local scripts.\n
  • Sanitization: Extracted markdown content is not escaped or validated before being included in the prompt output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 03:44 PM
Security Audit — agent-trust-hub — brand