interrogate
- Use it when:
- You want a change or design stress-tested by more than one model before it ships.
- It does:
- Restates the intent, gives each adversary its own worktree with your uncommitted changes, and keeps a finding only if rerunning its command shows a failure that breaks the intent.
- You get:
- The intent, each verified finding with its command, output, impact and who found it, and counts of dropped findings with reasons.
- Doesn't:
- Apply fixes, or report style, naming, structure or risks that need a future change to matter.
- Runs:
- Only when you call it.
Skill 5 of 14 in the pstack bundle, after arena and before blast-radius.
Only works inside its bundle: Runs the arena skill's runner setup, fan-out and verify steps; it also states the intent with in-your-own-words, which comes with the plugin but isn't listed here. Install pstack instead.
Written for Cursor's subagents and model names; the pstack plugin for Claude Code installs an adapted copy.
TL;DR
The scope is what you name, or else the branch's diff from its merge base through the working tree. It states the intent from your message, the commits and the PR description with pstack's in-your-own-words skill, and asks you when the intent is unclear. It then reuses the setup and fan-out of pstack's arena skill, with adversaries taken from the interrogate reviewers line of pstack's model rule. Each is told to prove every finding with a run against the real code, such as a failing test, a repro script or command output, to run a prototype or probe the tree for a design or plan, and that no findings is a valid result. Unlike arena, nothing is judged or picked: different adversaries find different failures, so the report keeps every verified finding, merging those that share a root cause.
Summarizes the skill at 8ecfb19.
npx skills add https://github.com/cursor/plugins --skill interrogate skills.sh pageSource on GitHubAll skills from cursor/plugins
- Total Downloads
- 5,300 on skills.sh, Nov 9, 2026
- Repo stars
- 15,776 cursor/plugins
- First seen
- May 23, 2026 on skills.sh
Weekly values
| Week to | Installs |
|---|---|
| Sep 21 | 287 |
| Sep 28 | 346 |
| Oct 5 | 392 |
| Oct 12 | 365 |
| Oct 19 | 424 |
| Oct 26 | 465 |
| Nov 2 | 426 |
| Nov 9 | 293 |
| Category |
Find root causes, review code and test what you built. Everything for building software: planning, coding, debugging, design, frameworks, cloud, agents and skills. |
|---|---|
| Type |
A multi-step process the agent follows: interview, debug, run a TDD loop, review. |
| Effect |
Its own steps create or change files, git state or local tools. |
| Needs |
Relies on the agent being able to spawn subagents. |
| Publisher |
Everyone else. Unverified vendor claims default to community. |
| Topics |
SKILL.md
From cursor/plugins at 8ecfb19. Checked daily, last changed Nov 6, 2026.
Frontmatter
name: interrogate
description: "Use for \"interrogate\", \"adversarial review\", \"multi-model review\", \"challenge this\", \"stress test this code\", \"find blind spots\", or \"tear this apart\". Adversaries try to break a change or design, and only findings proven by a run survive."
disable-model-invocation: trueInterrogate
Adversaries try to break a change or a design. A finding survives only when a run proves it. Report the findings. Do not apply fixes.
Frame
The scope is what the user names. Otherwise it is the branch diff from its merge base with the base branch, through the working tree. State the intent with the in-your-own-words skill, from the user's message, the commits, and the PR description. Ask the user when the intent is unclear.
Fan out
Run Phase A steps 3 and 4 and Phase B of the arena skill, with four changes. Read the interrogate reviewers line in place of arena runners. Copy the scope's uncommitted changes into each worktree. Spawn each adversary as generalPurpose. In place of the arena task and rationale, give each adversary the scope, the intent, and this brief:
Break this change. Prove each finding with a run against the real code, such as a failing test, a repro script, or command output that shows the wrong behavior. For a design or plan, run a prototype or probe the current tree. Keep repros in your own worktree. Report each finding with a title, the exact command, the output it printed, and the output the intent requires. Leave out anything you did not run. That covers style, naming, structure, and risks that need a future change to matter. No findings is a valid result.
Verify
Verify each finding per arena Phase F by rerunning its command. Keep a finding when the output shows the failure and the failure breaks the intent. Drop the rest. Merge findings that share a root cause.
Skip arena Phases C, D, and E. Adversaries find different failures, so the verdict keeps every verified finding, not one adversary's list.
Output
- Intent: the restatement.
- Findings: for each verified finding, the command, its output, how it breaks the intent, and the adversaries that found it. Say "no verified findings" when none survive.
- Dropped: for each adversary, the count of dropped findings and why.