| name | claudex-second-opinion |
| description | Suggests a cross-model second opinion after substantial or risky code changes. Use when a significant implementation, refactor, security-sensitive change, or uncertain bug fix has just been completed and is about to be committed or shipped. Advisory only — it recommends running /claudex:verdict; it never invokes Codex on its own. |
ClauDex Second Opinion
Overview
When a change is substantial or risky, suggest a cross-model review with
/claudex:verdict. Two frontier models trained by rival labs have different
blind spots — when both flag the same line it's the strongest free review
signal there is, and where they disagree is exactly where a human should look.
Hard rule: never run codex yourself under this skill. A cross-model
review spends the user's Codex quota and adds real latency. Explicit
invocation is consent — your job here is one well-timed suggestion, nothing more.
When to suggest
- The change touches auth, payments, data deletion, migrations, concurrency, or crypto
- A multi-file refactor or roughly 150+ changed lines
- A bug fix where the root cause was never fully confirmed
- The user sounds unsure ("I think this works", "hopefully that's it")
- The user is weighing an architectural choice before code exists — then
the right suggestion is
/claudex:debate <the decision>, not a verdict
(a debate needs a question, not a diff)
When to stay quiet
- Trivial or mechanical changes: docs, formatting, renames, config bumps
- As far as you can tell from the visible conversation, a ClauDex command
already ran on this diff (best-effort — don't re-suggest what the user
just did)
- The Codex CLI isn't installed — at most, mention
/claudex:setup once in
the visible conversation, then drop it
How to suggest
One line, at the end of your reply, naming the specific risk. Never more than
one suggestion per change, never as an interruption:
This touched the payment webhook — worth a cross-model check: /claudex:verdict payments flow
