ChatGPT for marketing, without building the process yourself
Founders are already using ChatGPT for marketing, one prompt at a time. Tin turns that into a system: open-source workflows with live data, schedules, review steps and memory, run by the agent that already knows your product.
codex mcp add tin --url https://app.tin.computer/mcpStart from what it already knows
ChatGPT knows your business from your projects and past conversations, so it can describe the product, the buyer and your constraints without a briefing. For anything that needs the code, connect GitHub in Tin and the workflows read the repository themselves.
Beyond marketing prompts
Most advice on ChatGPT for marketing is a list of prompts. A prompt cannot crawl your site, buy keyword data inside a cap, send an approved campaign from your mailbox or open a pull request. Tin gives ChatGPT those as bounded workflows with review steps, and keeps the results in a project it can read next week.
Organic growth
Audit, research, plan, draft and deliver, with fixes as pull requests.
Outreach
A shortlist from your own mailbox, and campaigns that send only what you approved.
Know your product the way a new user meets it
An AI tester signs up, uses every feature and reports what breaks. Ask ChatGPT to turn the report into tickets, or hand it to Codex to fix.
Creative
A brand character and short-form demo videos recorded from your live product.
Keep the record
Everything lands in project files and memory, and a weekly brief tells you what moved.
What every workflow here comes with
Battle tested on real products
These are the workflows we run for real companies. Claw Messenger went from about $45 to $2,105 in monthly revenue in 124 days on them. Each one has declared inputs, bounded execution and tests, and anything public waits for your review.
See the case study →Open source: read it, fork it, self-host it
Apache 2.0. Every workflow is a folder in the repository: the steps it runs, the data it reads and what it is allowed to do. Change it, or run the whole system on your own machine.
Open the repo →Runs on the agent that already knows your product
Tin connects to Claude Code, Codex or Cursor over MCP. That agent has your repository open and knows your product, customers and voice, so it fills the inputs and reviews the output. You never brief a new tool from zero.
All workflows →Questions
How do I add Tin to ChatGPT?
Add Tin as a custom connector (a remote MCP server) in ChatGPT's settings with the URL https://app.tin.computer/mcp, then sign in. Custom connectors depend on your ChatGPT plan and may need developer mode. If you work in a terminal, the Codex guide has a one-line install.
Is this proven on a real business?
Claw Messenger went from about $45 to $2,105 in monthly recurring revenue in 124 days with the SEO, outreach, site fixes and weekly brief run by the agent. The case study is on the homepage.
What does Tin cost?
The software is open source under Apache 2.0 and can be self-hosted, where you pay your own providers. The hosted app uses credits.
Does this work with other agents?
Yes. Tin is an MCP server, so Claude Code, Codex, Claude, Cursor and any other agent that speaks MCP can run the same workflows on the same projects.
Is Tin really open source?
Yes. Tin is licensed under Apache 2.0 and the source is at github.com/tin-computer/tin. You can read every workflow, change it, and self-host the whole system.
Which coding agents can run Tin workflows?
Any agent that speaks MCP, including Claude Code, Codex and Cursor. There is also an API, and the same projects run in the browser at app.tin.computer.
Why run marketing from a coding agent?
Because it already knows your product. It has the repository open, so it fills workflow inputs accurately, reviews output against what the product really does, and can act on the results in the same session.