Codex has been incrementally shipping hook support (PreToolUse, PostToolUse,
UserPromptSubmit, SessionStart, Stop), but the Codex converter silently
dropped all hooks during conversion -- the only converter that didn't even
warn. This adds partial hook conversion for the 5 compatible events and
specific warnings for everything unconvertible.
- Add CodexHookEventName union, CodexHookCommand, CodexHookMatcher, CodexHooks
types and hooks field to CodexBundle
- Add CODEX_EVENTS map with toolScoped flag, isBashCompatibleMatcher() helper,
and convertHooksForCodex() to the converter
- Write .codex/hooks.json with backup support in the writer
- Auto-enable [features] codex_hooks = true in config.toml when hooks present
- Refactor renderCodexConfig() to options object for extensibility
- Update Codex spec with hooks documentation
- 18 new tests (12 converter + 6 writer)
PostToolUse conversion depends on openai/codex#15531 merging.
Research agents (repo-research-analyst, git-history-analyzer,
best-practices-researcher, framework-docs-researcher) were using
shell commands like find, rg, cat, and chained pipelines for routine
codebase exploration. This triggers permission prompts in Claude Code
and degrades the user experience when these agents run as sub-agents.
Updated all research agents with platform-agnostic tool selection
guidance that prefers native file-search/glob, content-search/grep,
and file-read tools over shell equivalents. Shell is now reserved for
commands with no native equivalent (ast-grep, bundle show, git).
Git-history-analyzer additionally limits shell to one simple git
command per call with no chaining or piping.
Added tool selection rules to AGENTS.md so future agents follow
the same pattern by default.
Trevin Chow