agent-runtime-hooks

name	agent-runtime-hooks
description	Agent 运行时生命周期钩子 - 初始化/前置/后置/清理钩子，任务拦截链和事件总线
version	0.2.0
status	active
category	devsecops
tags	["agent","hooks","lifecycle"]

agent-runtime-hooks

Overview

Agent 运行时生命周期钩子 - 初始化/前置/后置/清理钩子，任务拦截链和事件总线

Category

devsecops

When to Use

Trigger this skill when the user requests operations related to agent runtime hooks.

Workflow

Pre-flight

1. Verify prerequisites are met (dependencies, credentials, environment)
2. Check configuration via agent-runtime-hooks-setup or equivalent if available
3. Validate user input parameters

Execution

1. Parse and validate the user's request
2. Execute the core operation using appropriate tools:
   • exec: Run CLI commands, scripts, or API calls
   • web_fetch: Call REST APIs when applicable
   • browser: Handle web-based flows when needed
3. Handle errors gracefully with clear messages

Post-execution

1. Format results for readability
2. Report success/failure with actionable next steps
3. Log operations if tracking is enabled

Key Constraints

• Always validate inputs before execution
• Handle authentication/authorization errors explicitly
• Respect rate limits and API quotas
• Provide Chinese-language output by default (可切换英文)
• Never expose secrets or tokens in output

Examples

Example 1: Basic Operation

User: 用 agent-runtime-hooks 执行基本操作
Assistant: [调用 exec 执行相应命令，返回格式化结果]

Example 2: Error Handling

User: 用 agent-runtime-hooks 操作一个不存在的资源
Assistant: [返回明确的错误信息，建议下一步操作]

Dependencies

• OpenClaw runtime >= 1.0
• Category-specific external tools (see individual commands)

Related Skills

• Check skills-audit for available skill registry
• Check category peers under devsecops