Skip to content
TrustedSkillsbeta
Back to all skills

Adversarial Review

Checked
poteto·vlatest·other·MIT

Simulates harsh, critical reviews to identify product weaknesses and improve robustness against negative feedback.

Repository

Install on your platform

Showing Claude Code. One command installs this skill into any of the agents below.

  1. 1

    Run this in your project

    terminal
    $ npx skills add poteto/noodle --skill adversarial-review -a claude-code

  2. Uses the open-source skills CLI and needs Node.js. Add -g to install it for your user instead of this project.

What we know about this skill

Publisher

poteto

Taken from the source listing. We haven't verified who controls this account.

Install target

Live repository

Not pinned. Installing fetches whatever the repository holds at the time you run the command.

No human has reviewed this code. The checks below are a static scan of the files at one commit — nobody has run this skill or judged whether it works. A skill runs with whatever access you give your agent, so read the source before you install it.

Automated safety checks

scanned

This skill passed all 6 checks. We read its files at one commit and looked for the things that make a skill dangerous to run — not for whether it works.

  • Declares what it does: passed

    SKILL.md parses and declares a name and a description.

    SKILL.md exists, its frontmatter parses, and it declares a name and a description.

  • No undeclared network calls: passed

    Makes no network calls.

    Every host the skill dials — from its scripts, its config endpoints or the commands it tells the agent to run — is either well-known package and API infrastructure or declared in its own frontmatter. Links in documentation are not counted.

  • No obfuscated payloads: passed

    No encoded or obfuscated payloads.

    No base64, hex or compressed blob is decoded and then executed, and no encoded shell commands.

  • No credential access: passed

    Reads no keys, tokens, keychains or credential files.

    Neither the code nor the SKILL.md instructions read SSH keys, cloud credentials, keychains, token stores, browser cookie databases or dotfiles that hold secrets.

  • No pipe-to-shell installers: passed

    No pipe-to-shell or download-and-run installers.

    No `curl … | sh` pattern, in the code or in the instructions, that runs code fetched at run time.

  • Pinned to a commit: passed

    Scanned poteto/noodle at 82d2921c52.

    The exact upstream commit SHA that was scanned is recorded, so the result refers to specific bytes.

Commit
82d2921c52
Manifest
.agents/skills/adversarial-review/SKILL.md
Scanned
4 files, 172 lines
This is a static scan, not a review. It reads the skill's files; it does not run them, and it cannot tell you whether the skill is any good or whether the code does what its description says. A skill runs with whatever access you give your agent.
What each check looks for →

About this skill

What it does

This skill simulates harsh, critical reviews of work (like code or text) to identify weaknesses and improve robustness against negative feedback. It spawns reviewers using a separate AI model ("opposite model") which challenges the original work from distinct perspectives grounded in cognitive science principles. The final output is a synthesized verdict; this skill is designed for review and analysis, not automatic changes.

When to use it

  • Reviewing code diffs or other significant changes before merging or deployment.
  • Identifying potential vulnerabilities or weaknesses in written content.
  • Gaining insights into how users might negatively perceive a product or feature.
  • Improving the overall quality and robustness of work by proactively addressing criticism.
  • Assessing whether a piece of work effectively achieves its intended purpose.

Key capabilities

  • Reviewer Spawning: Creates reviewers using a separate AI model (Codex or Claude) to provide adversarial feedback.
  • Lens-Based Review: Reviewers operate from distinct perspectives defined by "lenses" (Skeptic, Architect, Minimalist).
  • Intent Assessment: Requires explicit definition of the work's intended purpose before review begins.
  • Scalable Reviewer Count: The number of reviewers spawned depends on the size of the change being reviewed (1 for small changes, 2 for medium, and 3 for large).
  • Verdict Synthesis: Combines individual reviewer outputs into a single, synthesized verdict.

Example prompts

  • "Review this code diff: [code diff]"
  • "Assess these changes to the product description: [product description]"
  • "Provide an adversarial review of this draft document, focusing on clarity and potential misunderstandings."

Tips & gotchas

  • Model Dependency: Reviewers must be spawned using the CLI of the opposite AI model (Codex for Claude agents, Claude for Codex agents). Using subagents or internal delegation will not work.
  • Intent is Key: Clearly state the intended purpose of the reviewed work before initiating the adversarial review process.
  • No Automatic Changes: This skill provides a verdict; it's up to you to evaluate and decide which findings warrant action.