Back to blog
IA & LLMAI agentsCSRFtechnique

ChatGPT AgentForger: one phishing link deploying a rogue AI agent with employee access

Published on 2026-07-246 min readCleanIssue

> In short: AgentForger is a CSRF vulnerability in OpenAI's ChatGPT Workspace Agents, disclosed by Zenity Labs. A single phishing link — a benign-looking ChatGPT URL — could silently build, authorize, and deploy an autonomous AI agent inside the victim's organization, running with a real employee's access and its approval controls switched off. Fixed by OpenAI on June 8, 2026.

Why this matters to you

Most SaaS vendors we work with have either shipped an AI agent feature or are about to. ChatGPT Workspace, Copilot, and similar tools are now embedded in daily work — reading documents, calling APIs, sending messages on behalf of employees. AgentForger shows what happens when an old web vulnerability class (CSRF) meets a new high-privilege surface (an autonomous agent): the blast radius is no longer a forged blog comment, it's a rogue insider that runs itself.

If your product lets users create, configure, or authorize AI agents through a web UI, the same pattern applies to you.

The flaw in two sentences

ChatGPT Workspace Agents are built and authorized through web requests that the Agent Builder processes in the user's browser session. Because the state-changing requests were not protected against CSRF, an attacker could craft a link that — when clicked by a signed-in employee — caused the Agent Builder to stand up a new agent, configure it with the employee's existing permissions, and disable the approval prompts that would normally gate risky actions.

The result is an attacker-controlled AI agent living inside the company's trust boundary, doing the attacker's bidding with the employee's access. No password theft, no token replay, no malware — just one click on a link that looks like a normal ChatGPT URL.

Why the impact is larger than a normal CSRF

A traditional CSRF forges one action: transfer money, change an email, post a comment. An agent-deployment CSRF forges an *autonomous worker* that then performs many actions over time, with the victim's identity, until it is noticed and removed. Three properties make it acute:

  • Persistence. The agent keeps running after the phishing page is closed. The attacker does not need to maintain a session.
  • Inherited trust. The agent is created under the employee's identity, so downstream systems see legitimate activity.
  • Disabled approvals. Switching off the human-in-the-loop gates is what turns a noisy action into a silent one.
  • What to do

  • On ChatGPT Workspace specifically: nothing — OpenAI fixed it on June 8. Review any agents that were created around that window if you were a target.
  • If you build an AI agent feature, treat every state-changing endpoint (create, authorize, disable approvals, attach a tool) as CSRF-protected by default. Use anti-CSRF tokens or SameSite=strict cookies, not origin checks alone.
  • Separate "deploy" from "authorize." Building an agent and giving it production access should be two distinct user actions, each requiring an explicit interaction. AgentForger worked because one link did both.
  • Make approval-gating un-disable-able for high-risk tools. If an agent can call billing, send external email, or delete data, the approval gate should be a server-side control that the agent creator cannot turn off via the UI.
  • Add an agent inventory and anomaly detection. You should be able to list every agent running in your tenant, who created it, and what it can do — and alert on agents created outside expected patterns.
  • The broader point

    AgentForger is a preview of the next decade of SaaS attacks. As soon as a product lets users create autonomous workers that act on their behalf, every old web vulnerability — CSRF, IDOR, XSS, SSRF — gets a multiplier on impact. The defense is not a new control; it is applying the old controls (anti-CSRF, least privilege, separate consent) to the new agent-creation surface with the seriousness the blast radius demands.

    Building HR, payroll, or recruiting software? CleanIssue performs security audits for HR SaaS in real-world conditions, no source code access needed. For a first read of your exposure, start with an external review of your application.

    Sources

    Written by CleanIssue
    Reviewed on 2026-07-24

    Related services

    If this topic maps to a real risk in your stack, these are the most relevant CleanIssue audits.

    Need an external review of your HR SaaS?

    Share your product, stack, and client context. We will come back with the right review scope.

    Discuss your audit