hooks

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides examples of shell commands used within hook configurations, such as using pnpm prettier for automatic code formatting. These examples are documented for automation purposes and utilize standard development tools.
  • [INDIRECT_PROMPT_INJECTION]: The skill outlines a pattern for hooks to process agent tool inputs to validate or block potentially dangerous actions.
  • Ingestion points: SKILL.md (example bash script reading tool_input via stdin).
  • Boundary markers: Not applicable; the examples demonstrate logic for parsing and inspecting inputs.
  • Capability inventory: Shell command execution within the hook environment.
  • Sanitization: The provided example includes regex-based validation to identify and block destructive commands like 'rm -rf /' before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 09:27 AM
Security Audit — agent-trust-hub — hooks