create-hook

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPERSISTENCEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to generate and execute shell and Node.js scripts as part of a 'Test & validate' workflow to ensure newly created hooks function correctly.
  • [PERSISTENCE]: The skill is designed to create executable scripts in ~/.claude/hooks/ or .claude/hooks/ and modify settings.json to register these scripts. This allows for automated code execution at specific lifecycle points in future sessions.
  • [PRIVILEGE_ESCALATION]: The workflow includes granting execution permissions to generated scripts using chmod +x to ensure they can be invoked by the platform's hook runner.
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes external project data to make decisions:
  • Ingestion points: Project configuration files including package.json, tsconfig.json, .prettierrc, and .eslintrc.*.
  • Boundary markers: Absent; the skill scans for patterns without explicit delimiting of external data.
  • Capability inventory: The skill has the ability to write files, create directories, modify permissions, and execute shell commands.
  • Sanitization: Instructions do not specify sanitization or validation of the content discovered in project configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:10 AM
Security Audit — agent-trust-hub — create-hook