Back to Marketplace

DECEPTICON

Open-source pickSecurity & Compliance

Dark-pattern robustness evaluation for web agents, based on the DECEPTICON benchmark and methodology.

By Phil Cuvin, Hao Zhu, Diyi Yang / RAPR AIv1.0.0Package license: MITResearch Workflow

DECEPTICON package details

securityevaluationweb-agentsdark-patternsqa

Open-source curation scope

This listing is a convenience wrapper for discovery, attribution, setup links, install commands, and agent guidance. This package curates public setup links, install commands, and agent instructions. It does not bundle or relicense the upstream project unless explicitly stated. Users can also go directly to the public upstream source linked on this page.

How to get started

Install RAPR AI

Download and install RAPR AI on your computer

Find in Marketplace

Open RAPR AI, go to Packages, and browse the marketplace

Install from Marketplace

Click Install. RAPR sets up the wrapper package, connector guidance, or skill instructions for this listing.

DECEPTICON

A security and evaluation skill inspired by the DECEPTICON research on how dark patterns manipulate web agents. Use it to test whether browser agents, QA flows, and web automation tasks can be steered into outcomes that conflict with the user's intent.

What you can do

  • Review web tasks for dark-pattern risk.
  • Design adversarial QA for browser agents.
  • Audit agent traces for unwanted purchases, subscriptions, permissions, or account changes.
  • Create guardrails around consent and irreversible actions.

This is a defensive evaluation workflow. Do not create manipulative UI.

Ready to try DECEPTICON?

Download RAPR AI and connect DECEPTICON in seconds.