feat(plugin): enhance context management and skill invocation (#239 )

* fix(shellcheck): declare and assign separately in router.sh (SC2155)

* feat(plugin): enhance context management and skill invocation

- Update .context-manifest.json to include profile and detailed source information.
- Revise context-scout.md to clarify context root discovery and navigation.
- Modify commands to replace deprecated skills with updated ones (brainstorming → approach, systematic-debugging → debugger).
- Streamline install-context.md for clarity and remove interactive profile selection.
- Improve oac-status.md to provide a more comprehensive status report.
- Add OAC system paths to session-start.sh for better context management.
- Remove obsolete skills and improve documentation for existing commands.

2026-02-23 12:12:29 +00:00

1.7 KiB

Raw Blame History

Hooks System

Purpose: User-defined shell commands that execute at Claude Code lifecycle points

Core Concept

Hooks provide deterministic control over Claude Code's behavior. Unlike prompting (which suggests), hooks guarantee certain actions happen at specific points in the workflow.

Key Points

Hooks are shell commands triggered by events
Run automatically during agent loop with your credentials
Can block operations (exit code 2), allow silently, or provide feedback
Configure in settings.json or plugin's hooks/hooks.json
Matchers filter which tools/events trigger the hook

Common Use Cases

Use Case	Event	Example
Auto-format	`PostToolUse`	Run prettier after Edit/Write
Notifications	`Notification`	Desktop alert when input needed
File protection	`PreToolUse`	Block edits to .env files
Logging	`PreToolUse`	Track all bash commands
Feedback	`PostToolUse`	Lint check after code changes

Quick Example

{
  "hooks": {
    "PreToolUse": [{
      "matcher": "Bash",
      "hooks": [{
        "type": "command",
        "command": "jq -r '.tool_input.command' >> ~/.claude/bash-log.txt"
      }]
    }]
  }
}

Security Warning

Hooks run with your credentials. Malicious hooks can exfiltrate data. Always review hook implementations before adding them.

../lookup/hook-events.md - All hook events
../examples/hooks/ - Hook examples

Reference: https://docs.anthropic.com/en/docs/claude-code/hooks

1.7 KiB Raw Blame History