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。

All Hooks

插件 活跃

Complete collection of 28 automation hooks for event-driven workflows

目的

To extend Claude Code's capabilities with event-driven automation and notifications, allowing users to streamline workflows and stay informed about project changes.

功能

  • Bundled collection of 28 automation hooks
  • Event-driven automation for git, file edits, and tool usage
  • Notification hooks for Slack, Discord, and Telegram
  • Hooks for testing, security scanning, and code formatting
  • Marketplace for discovering other Claude Code extensions

使用场景

  • Automate git staging and commits after file edits
  • Receive notifications on Slack or Discord when Claude Code finishes tasks
  • Automatically run tests or linters after code modifications
  • Enhance security by scanning for vulnerabilities and protecting critical files

非目标

  • Replacing core Claude Code functionality
  • Providing complex, multi-step agentic workflows within a single hook
  • Managing cloud infrastructure or complex deployment pipelines directly

Security

  • warning:Secret ManagementSeveral hooks (e.g., Slack, Discord, Telegram) require webhook URLs or bot tokens as environment variables, but these are not explicitly handled via `userConfig` with `sensitive: true` in the `plugin.json`, potentially leading to plain text storage in settings.
  • warning:Hook securityNetwork-calling hooks for Slack, Discord, and Telegram do not have a documented disable toggle or environment variable to turn them off per-session without uninstalling the plugin.
  • warning:Keychain-stored secretsSecrets like `SLACK_WEBHOOK_URL` and `TELEGRAM_BOT_TOKEN` are expected to be set as environment variables, but there's no indication they are routed through `userConfig` with `sensitive: true` for secure storage in the OS keychain.

Code Execution

  • warning:ValidationWhile some scripts check for the presence of environment variables, there is no explicit schema validation or sanitization for input parameters like file paths or commands within the hooks, beyond basic checks.
  • warning:Error HandlingSome scripts use `|| true` or `2>/dev/null` to suppress errors, and error reporting is inconsistent, lacking structured error fields for the agent.
  • info:LoggingThe `performance-monitor` hook logs to a local file, and some hooks redirect stderr, but a comprehensive, structured audit log for all actions is not consistently implemented across all hooks.

Compliance

  • info:GDPRThe hooks send notification data to external services, but this requires user-provided credentials and focuses on operational events, not personal data submission without consent.
  • warning:Hook privacyNetwork-calling hooks send notification data to external services (Slack, Discord, Telegram) which requires user-provided credentials and is enabled by default without explicit opt-in or a clear mechanism to disable them per session.

Scope

  • warning:Tool surface sizeThe 'all-hooks' package bundles 28 distinct hooks, which is significantly more than the target range of 3-10, potentially leading to complexity in selection.

Invocation

  • warning:Overlapping near-synonym toolsThere are multiple hooks for similar notification types (e.g., multiple Slack/Discord/Telegram variations for error vs. general stop notifications), which could lead to confusion.
  • warning:Hook matcher tightnessSeveral hooks use a wildcard matcher `*`, which is too broad and could lead to unintended executions or performance impacts.

Errors

  • warning:Actionable error messagesError messages are often generic or suppressed (e.g., `2>/dev/null || true`), lacking clear root cause framing and remediation steps for the user.

安装

请先添加 Marketplace

/plugin marketplace add davepoon/buildwithclaude
/plugin install all-hooks@buildwithclaude

质量评分

75 /100
1 day ago 分析

信任信号

最近提交5 days ago
星标2.9k
许可证MIT
状态
查看源代码