跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

Hookify Plugin

插件 活跃

Easily create hooks to prevent unwanted behaviors by analyzing conversation patterns

1 个 Skill 0 个 MCP
目的

Empower users to proactively prevent unwanted or risky behaviors by easily defining custom rules based on conversation analysis and explicit instructions.

功能

  • Create custom hooks via markdown files
  • Analyze conversations for unwanted behaviors
  • Define rules for bash, file, stop, and prompt events
  • Block or warn on rule matches
  • Interactive rule management commands

使用场景

  • Preventing the use of dangerous bash commands like `rm -rf`
  • Warning about sensitive information in file edits
  • Ensuring best practices like running tests before stopping
  • Enforcing coding standards by detecting debug statements

非目标

  • Replacing core Claude Code functionality
  • Managing external dependencies
  • Providing a full-fledged security auditing tool

工作流

  1. User invokes `/hookify` command
  2. Plugin analyzes conversation or uses provided instructions
  3. User selects behaviors and actions (warn/block)
  4. Plugin generates `.claude/hookify.{rule-name}.local.md` files
  5. Hooks dynamically load rules on next tool use
  6. Plugin enforces defined rules during agent execution

实践

  • Rule definition
  • Behavioral analysis
  • Workflow enforcement

先决条件

  • Python 3.7+

Trust

  • warning:Issues AttentionWith 9462 issues opened and 22831 closed in the last 90 days, the closure rate is approximately 71%, but the high volume of open issues suggests potential delays in maintainer response.

安装

请先添加 Marketplace

/plugin marketplace add anthropics/claude-code
/plugin install hookify@claude-code-plugins

质量评分

95 /100
2 days ago 分析

信任信号

最近提交2 days ago
星标123.1k
许可证MIT
状态
查看源代码