OpenClaw Hook Examples

Provides example hooks for OpenClaw to intercept, modify, block tool calls, switch models, log actions, and handle subagent events.

This file is a set of examples, not an active plugin: installing it as-is does nothing because there is no install or code to run. If you or a developer copy these examples into a real plugin, be aware: hooks can block or modify tool calls and may log parameters and results — avoid logging or exporting sensitive data, validate and sanitize inputs carefully, and confirm any model/provider strings (e.g., ollama) match your environment and credential posture. Also fix the minor inconsistencies (sync vs async, intended log destination) before using in production. If you plan to run a plugin built from these examples, review where logs go and which sessions are considered "elevated" to prevent accidental exposure or overly broad blocking behavior.

Type: OpenClaw Skill Name: openclaw-hook-examples Version: 1.0.0 The bundle contains educational documentation and code examples for OpenClaw hooks in SKILL.md. The examples demonstrate defensive security practices, such as auditing tool usage, blocking dangerous shell commands (e.g., 'rm -rf' and fork bombs), and implementing permission checks for sensitive tools. No malicious logic, data exfiltration, or prompt injection attempts were found.