v1.4.1
Patch release correcting the per-model effort default that v1.4.0 introduced.
Stop defaulting reasoning effort per model
v1.4.0 gave every friendly alias a high default effort. That duplicated a catalog the host CLI owns — exactly like the pinned model IDs removed in that same release — and it overrode your own Claude Code effort setting. If you had "effortLevel": "xhigh" in ~/.claude/settings.json, the plugin silently downgraded you to high.
--effort is now forwarded only when you pass it, so each model keeps whatever effort Claude Code defaults to, and Claude Code stays the authority on which levels a model supports. --model still defaults to opus.
If you want a specific effort level: pass --effort xhigh per invocation, or set effortLevel in your Claude Code settings — the latter now actually takes effect.
Blank flag values no longer crash
resolveModel and resolveEffort both return undefined for whitespace-only input, and both were pushed into the spawn argv unconditionally, failing with ERR_INVALID_ARG_TYPE. Both flags now gate on the resolved value and drop blanks like any other unset flag.
Full changelog: https://github.com/sendbird/cc-plugin-codex/blob/main/CHANGELOG.md