Yeachan-Heo/oh-my-claudecode v3.8.2

v3.8.2 - Fix hooks API output format

on GitHub

Bug Fixes

• hooks: Use hookSpecificOutput.additionalContext instead of invalid message field (#205)

  All hook scripts were using a message field in their JSON output which is not part of the Claude Code hooks API. This caused hook messages to be silently dropped — keyword detection, skill injection, session restore, pre-tool reminders, and post-tool verification were all non-functional.

  Fixed 5 scripts to use the documented format:

  • keyword-detector.mjs (UserPromptSubmit) - 16 instances fixed
  • skill-injector.mjs (UserPromptSubmit) - 1 instance fixed
  • session-start.mjs (SessionStart) - 1 instance fixed
  • pre-tool-enforcer.mjs (PreToolUse) - 1 instance fixed
  • post-tool-verifier.mjs (PostToolUse) - 2 instances fixed

Documentation

• Added RULE 5 (Documentation-First Development) to CLAUDE.md
• Requires consulting official documentation before implementing with SDKs/frameworks/APIs
• Prevents future assumptions about undocumented API behavior

Upgrade

After upgrading, restart Claude Code for the hook changes to take effect (hooks are snapshotted at startup).