---
name: interrogate
description: "Use for \\\"interrogate\\\", \\\"adversarial review\\\", \\\"multi-model review\\\", \\\"challenge this\\\", \\\"stress test this code\\\", \\\"find blind spots\\\", or \\\"tear this apart\\\". Multiple LLM reviewers challenge changes from independent angles."
---

# Interrogate

Spawn independent reviewers to adversarially review code changes. Actual model diversity can add signal, but the default review passes inherit the parent. Disclose that these are independent same-provider reviews. Assigned personas do not create model diversity. The deliverable is a synthesized verdict. Do NOT auto-apply changes.

## Step 1, Scope

Identify what to review from context:

- If the user points at specific files or a diff, use that.
- If on a feature branch, run `git diff main...HEAD` (or the appropriate base branch) for the full changeset.
- If the user's message references recent work, gather the relevant files.

Package the diff or file contents with any surrounding context files the reviewers need to understand the code.

## Step 2, Intent

Before spawning reviewers, state the intent in one clear paragraph from the user's message, commit messages, the PR description if one exists, and the code. If you're unsure about the intent, ask the user before proceeding.

## Step 3, Spawn Reviewers

Read the model and delegation contracts in [CODEX.md](../../CODEX.md). Use two independent reviewers by default, Reviewer A and Reviewer B, with inherited parent settings. A supported explicitly chosen panel can change the seat count or available models. Launch within actual free slots, queueing the rest. Each receives the same scope and a read-only assignment, with no edits or external posts.

Read `references/reviewer-prompt.md` and fill in the same template for every reviewer with the stated intent, the diff or file contents, the rubric from `references/rubric.md`, and the code-quality lens from `references/code-quality-review.md`.

Report actual model evidence per seat. When exact model identity is not exposed, say `parent model inherited; exact model unavailable`. Do not claim cross-provider review from native OpenAI-only seats. A rejected model does not authorize changing setup, opening a PR, or selecting an unrequested replacement. Report the failure and use an inherited pass only when the request permits the assurance change.

## Step 4, Synthesize

As results come back, build a unified picture. Parse every reviewer's findings. Merge findings that describe the same issue differently, and note which models raised each one. Findings two or more models raised independently are the highest signal. Read a lone model's finding, but weight it accordingly. Note disagreements. If one model flags something and another explicitly says the opposite, that is context for the verdict.

## Step 5, Lead Judgment

You are the lead reviewer, a pragmatic senior engineer, not a neutral aggregator. Read `references/lead-judgment.md` for the full framework. Put every finding in one of four buckets, with the model(s) that raised it and a one-line rationale.

- **Act on**. Real issues affecting correctness, security, or maintainability given the actual goals. They would block a real PR.
- **Consider**. Legitimate, but you're not sure they outweigh the cost of addressing them now. Worth the user's attention.
- **Noted**. Technically valid but not actionable. Context-dependent, premature optimization, or low-impact at this stage.
- **Dismissed**. Wrong, nitpicky, or missing context, with a brief reason.

## Output Format

### Intent
> [The stated intent paragraph from Step 2]

### Reviewers
- Reviewer [label]: [model name], [N findings] (one bullet per reviewer)

### Act On
[Each: description, which models raised it, why it matters.]

### Consider
[Each: description, which models raised it, the tradeoff.]

### Noted
[Brief list.]

### Dismissed
[Each with a brief rationale.]

### Agreement Map
[Where models agreed and diverged, and what that pattern tells us.]