Skip to content

BumpyClock/agent-templates / skills/pstack/interrogate

interrogate

Unverified

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 one reviewer per configured model to adversarially review code changes. The adversarial signal comes from model diversity, not assigned personas. 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, compare the working tree with the merge base of the base branch and HEAD. Include staged and unstaged changes and relevant untracked files, not only committed changes.
  • 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. Give reviewers separate writable locations for reproduction scripts and fixtures. If they use separate checkouts, include the scoped uncommitted changes so they test the reviewed state.

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

Launch all reviewers in a single message with the Task tool, one per entry on the interrogate reviewers line in ~/.cursor/rules/pstack-models.mdc, labeled Reviewer A, B, and so on to match the entry count. If the rule or that line is missing, use the table defaults.

SubagentDefault model
Reviewer Aclaude-opus-5-5-xhigh
Reviewer Bgrok-4.7-xhigh-fast
Reviewer Cgpt-6.1-sol-high

Launch the model using the appropriate tool or cli as defined in ~/.agents/configured-models.md . If it rejects a table default, spawn with the closest equivalent from the valid slugs in its error message (prefer the same family and reasoning tier), and open a separate PR to update the default table. Do not block the review on the slug issue. Never treat an alias entry as a rejected slug or apply either fallback to it.

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. Pick each subagent's reasoning level for the job, per the Subagents section of the configured-models.

Step 4, Verify and synthesize

For each behavioral defect, require the exact reproduction command, actual output, and output required by the stated intent. Rerun the command against the reviewed state before treating the finding as verified. For a design or plan, use a bounded prototype or probe when it can test the claim. Keep reproduction changes isolated from the user's working tree. Do not execute a reproduction whose effects exceed the task's authorization.

Drop disproven findings and merge findings with the same root cause. Label findings that cannot be reproduced as unverified, with the missing prerequisite. Keep source-backed design and maintainability concerns in the local judgment categories below, but do not present them as runtime-proven defects. No verified findings is a valid result.

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, and verification status. For behavioral defects, include the reproduction command, actual output, and expected output.]

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.]

finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 318kUpdated Claude CodeCodexUnverifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 318kUpdated Claude CodeCodexUnverifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 318kUpdated Claude CodeCodexUnverifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 196kUpdated Claude CodeCodexVerifiedInstall