OpenAI-CodexAgent
OpenAI-CodexAgent is an autonomous browser agent operated by OpenAI. Instead of scraping content, it actually uses your site, clicking buttons, filling forms, completing checkouts, doing what a human user would do.
Traffic looks closer to a real user session than to a crawler. Pages are fully rendered, JavaScript runs, and interactions are sequential. The agent often runs on behalf of a paying customer who wants a specific outcome.
This is the leading edge of agentic commerce. The agents that buy your product, book your hotel room, or fill out your lead form on behalf of a human will live in this category.
See OpenAI-CodexAgent on your own site
Match the User-Agent header on incoming requests against the pattern below.
regex
For higher confidence, also verify the source IP against the operator's published ranges. UA strings can be spoofed; IP ownership is harder to fake.
Renders JavaScript
Yes
IP verification
Published IP ranges
Crawl frequency
Burst, user-driven
Honors robots.txt
Yes
Honors Crawl-delay
Varies
OpenAI runs 17 bots in total. Each one is a separate user-agent so you can allow or block them independently.
Agentic Browser
8- ChatGPT Agent
- ChatGPT-Agent
- OAI-Operator
- ChatGPT-Browser
- ChatGPT Atlas
- OpenAI-Operator
- OpenAI-CodexAgentYou are here
- OpenAI-Shop-Agent
Live-Fetch AI
2Agentic Commerce
2AI Search Index
1AI Coding Tool
1Link Unfurler
1Should I let OpenAI-CodexAgent through?
In most cases, yes. Agentic commerce is growing fast. Blocking these agents means losing the customers they act on behalf of. If volume gets noisy, rate-limit it before you block it outright.
Does blocking OpenAI-CodexAgent affect my Google rankings?
No. OpenAI-CodexAgent acts on behalf of one specific user at a time, not on behalf of a search index. Classical SEO is unaffected. The trade is whether you want users delegating tasks to OpenAI to be able to reach your site.
How do I confirm a request is really from OpenAI-CodexAgent?
Two checks. The User-Agent header should match a known OpenAI-CodexAgent string, and the request's source IP should fall inside OpenAI's published ranges. The User-Agent alone is trivially spoofable, so the IP check is what gives you confidence. OpenAI publishes the ranges so you can validate at the CDN or edge.
Can OpenAI-CodexAgent take actions on my site, like buying or signing up?
In principle yes. Agentic browsers can fill forms, click buttons, and complete transactions. That is the entire point. If your site leans on automation detection to prevent bots from acting, OpenAI-CodexAgent is a different threat model from a passive crawler and you should think about it separately.
How is OpenAI-CodexAgent different from OpenAI's other bots?
OpenAI splits work across multiple user-agents so site owners can decide on each one independently. Training crawlers, live-fetch agents, search indexers, and agentic browsers each get their own name. Worth scanning the rest of the OpenAI family above to see which ones actually matter for your site.
What's the cleanest way to control OpenAI-CodexAgent?
Two layers. Robots.txt for the polite crawlers that read it, and rules at your CDN or edge for the ones that don't. Rankly's Agent Experience handles both from a single config, so you can allow, block, rate-limit, or serve a stripped-down version per bot. Agent Analytics handles the observation half so you know which bots are actually worth a rule.
Verify everything above against the operator's own documentation.