OpenAI dots + Wayfern

Your dot asks. Codex calls Wayfern.

  • Through your connected computer
  • Codex reads ~/.codex/config.toml
  • Computer online, ChatGPT app open

A dot works for you all day, also while you sleep. OpenAI does not say if a dot can use a custom MCP plugin, so a dot alone has no Wayfern tools.

Connect your computer to the dot and add Wayfern to Codex there. The dot can then start a local Codex task that scrapes, searches and captures pages with Wayfern, and your custom rules decide when it must ask you.

Note:OpenAI does not say if a dot can use a custom MCP plugin. This page uses the route that OpenAI documents: a Codex task on your connected computer.

A dot in the cloud sends a job down to your computer. A Codex task there calls Wayfern to scrape five pricing pages, and one changed price goes back up to the dot, which tells you.

Setup

Give your dot a computer with Wayfern in Codex.

Do these steps on the computer that you connect to the dot.

  1. 1

    Add Wayfern to Codex

    Or run codex mcp add wayfern --url https://api.wayfern.com/mcp.

    ~/.codex/config.toml

    [mcp_servers.wayfern]
    url = "https://api.wayfern.com/mcp"
  2. 2

    Sign Codex in

    A browser opens on wayfern.com. Sign in and click Allow.

    bash

    codex mcp login wayfern
  3. 3

    Connect the computer

    In the ChatGPT app on this computer, open your dot and go to Computers. Find Your computer, click Allow access and confirm. A dot connects to one personal computer at a time.

  4. 4

    Set the rules

    Open Settings, Personalization, Custom rules. Set starting a Codex task to Ask before taking action until you trust the job.

  5. 5

    Give the dot the job

    prompt

    Every morning at 7, start a Codex task on my computer: "With the Wayfern tools, read the pricing pages in ~/watch/urls.txt and list each price that changed since yesterday." Tell me the result.

Other ways to connect

  • A token instead of OAuth

    For a computer where no one signs in. Create a token with the mcp scope on the Tokens page and put it in WAYFERN_API_TOKEN.

    ~/.codex/config.toml

    [mcp_servers.wayfern]
    url = "https://api.wayfern.com/mcp"
    bearer_token_env_var = "WAYFERN_API_TOKEN"
  • As an agent skill

    Codex loads skills from ~/.agents/skills. The skill calls the REST API with curl and needs WAYFERN_API_TOKEN in the environment.

    bash

    mkdir -p ~/.agents/skills/wayfern
    curl -fsSL https://wayfern.com/skills/wayfern/SKILL.md -o ~/.agents/skills/wayfern/SKILL.md

Route

A Codex task on your computer is the bridge.

The dot runs in OpenAI's cloud. A Codex task on your connected computer has Wayfern in its config, and OpenAI documents each step of that path.

Remote MCP OAuth sign-in
Best fit
Codex on the connected computer signs in with codex mcp login wayfern. A Wayfern plugin on the dot itself is not tested.
MCP + token Authorization header
Works
bearer_token_env_var in config.toml, for a computer where no one signs in.
Agent skill SKILL.md
Works
Codex loads SKILL.md from ~/.agents/skills. Codex's sandbox can block curl's network calls.

Example tasks

Jobs for an agent that does not sleep

  • 01

    Watch prices

    “Every morning at 7, read the pricing pages of our five competitors. Tell me only when a price or a plan changes.”

    scrape_page at 1 credit a page, through the Codex task.

  • 02

    Brief me on the news

    “At 6:30 each weekday, search Google and Bing for news about our company and our two main rivals from the last day. Send me five lines.”

    search_web bills 2 credits for each provider page.

  • 03

    Check our site at night

    “Each night, capture full-page screenshots of our home and pricing pages. Tell me in the morning if anything looks broken.”

    capture_screenshot costs 5 credits a capture.

  • 04

    Follow a changelog

    “Every Monday, crawl our payment provider's API changelog, 20 pages at most, and list the entries from last week.”

    crawl_site bills 1 credit for each page it scrapes.

Tools

What OpenAI dots can call

The same tools on every route, at the REST API rates. Every result reports credits_charged.

All tool inputs
  • scrape_page Fetch one page through an anti-detect browser and return clean markdown, HTML, links and metadata. The screenshot, pdf and json formats add +4 credits (once); summary adds +2. 1 credit / page
  • extract_data Extract structured data from one or more pages with a prompt and an optional JSON Schema. Re-running over pages already fetched is free. 1 credit / uncached source page
  • map_site List a site's URLs from its sitemaps and page links. 2 credits / 10 URLs
  • capture_screenshot Capture a clean image of a page: cookie banners and ads removed, lazy images loaded, tall pages captured end to end. Returns a short-lived download URL and an artifact key, not image bytes. Slices of a full page cost nothing extra; a fail_if_contains match is not charged. 5 credits / capture
  • brand_profile Identity, colours, logos, social links and company metadata for a domain. 10 credits / profile
  • site_styleguide Colours, typography, spacing, radii, shadows and the design tokens a site declares. 10 credits / site
  • site_fonts Font families, where each is served from, available weights, file formats and where they are used. 5 credits / site
  • crawl_site Start a bounded multi-page crawl. Returns a job id immediately and charges nothing up front; pages are billed as the crawl runs. 1 credit / successful page
  • crawl_status Poll a crawl started with crawl_site and read a page of its results. The pages were billed as they were crawled. Free
  • search_web Search Google, Bing, and DuckDuckGo. Returns ranked results, ads, provider failures, pagination, and related queries. 2 credits / provider page
  • list_search_engines List live provider ids and labels. Free

Good to know

Who can use dots
OpenAI lists Pro 100, Pro 200 and Pro 500 for users over 18 outside the EEA, the UK and Switzerland. Business Premium and Enterprise come next. The rollout is gradual.
Pending requests
Open your dot in the ChatGPT desktop app and go to Activity.
ChatGPT on the web
It does not read config.toml. The ChatGPT page has the plugin steps.