---
title: 'In the News: November 20, 2026 (Morning)'
description: "OpenAI's Performance Factory runs Codex agents from latency alert to verified fix, with a second agent assigned to disprove each diagnosis."
canonical_url: 'https://darkfactory.dev/news/2026-11-20-morning'
markdown_url: 'https://darkfactory.dev/news/2026-11-20-morning.md'
collection: news
date_published: '2026-11-20T07:05:00-05:00'
date_modified: '2026-11-20T07:05:00-05:00'
---

# In the News: November 20, 2026 (Morning)


OpenAI has published how its Performance Factory takes a latency alert through diagnosis, a proposed fix and a recovery check, including the review step it added after agents settled on wrong causes. A day earlier it documented the Codex sandbox, and one gap the post names is worth checking in your own config: stdio MCP servers start outside the sandbox.

## 1. OpenAI's Performance Factory sends a second agent to disprove each diagnosis

**[How OpenAI's Performance Factory keeps ChatGPT fast](https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/)** · Okay Zed and Cyrus Forbes, OpenAI · OpenAI Engineering, November 19, 2026

When a configured performance alert fires, the Regression Workflow opens a task. Codex agents look for the root cause and propose a fix as a pull request; an engineer reviews and deploys it, and another agent then checks whether performance returned to baseline. Task state is kept in a task service outside any agent's context. A dispatch loop assigns tasks that are ready, and subscriptions wake a task on events such as a review comment or a completed deployment. OpenAI says the workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months.

The post is direct about what went wrong. "Early on, agents sometimes stopped at the first plausible explanation they found, and some of those diagnoses turned out to be wrong." OpenAI added a separate Codex session that tries to disprove each diagnosis, and an adversarial review of each proposed fix as well. In a manual review of 100 root cause suggestions, it graded 70% as accurate. Fixes have been harder than diagnosis: some proposed changes were "more complicated than necessary or introduce new issues." Changes to skills or models are backtested against past incidents. The investigator cannot use saved memories, the incident writeup or earlier runs, and its transcript is checked afterward for anything that gave the answer away. All of these figures are OpenAI's own, and no outside check of them has been published.

**Why it matters:** The adversarial reviewer and the controls on backtests are the parts to copy. If an investigation agent can see saved notes or evidence gathered after recovery, a change to its skill can score better on past incidents than it will on the next one.

## 2. OpenAI documents the Codex sandbox and says stdio MCP servers run outside it

**[Inside the Codex sandbox](https://openai.com/index/inside-the-codex-sandbox/)** · Michael Bolin, OpenAI · OpenAI Engineering, November 18, 2026

Codex runs commands in the user's own environment and limits them with each platform's sandboxing: Seatbelt on macOS, bubblewrap with seccomp on Linux. The built-in `:workspace` profile allows writes inside the workspace but keeps `.git` and `.codex` read-only, allows broad reads, and turns network access off. The post names a gap in those defaults. An agent could inject code into a locally installed package such as `node_modules/prettier`, which Git often ignores, and an editor that runs that copy on save would execute it outside the sandbox.

MCP is a second gap: "For a stdio MCP server configured in config.toml, the Codex harness does not attempt to sandbox the command used to start the server." The post shows how to wrap such a server in `codex sandbox --permission-profile`, with network access off and writes limited to named directories. It also notes that broad wildcard domain rules or access to DNS services can still permit DNS-based exfiltration through the network proxy.

**Why it matters:** If your Codex setup starts stdio MCP servers, they run with your full user permissions unless you wrap them yourself. The `codex sandbox` command also lets you test a permission profile against the specific writes it is meant to block before an agent runs under it.
