TinTinWorkflows
GitHubSign in

Every technical SEO finding, fixed in one pull request you review

Tin takes a whole audit, checks each finding on the live site again, and fixes everything it can in one pull request: indexing, the sitemap and robots.txt, redirects, page structure, accessibility, structured data, social previews and internal links. Copy and steps outside the repository are listed, not guessed.

Try in browser →
codex mcp add tin --url https://app.tin.computer/mcp
Technical SEO fixes as one PR
organic.technical_fix
outputUnmerged GitHub PR and reports/technical-fix/{run_id}/RESULT.md
guardrailNever merges or deploys, and never writes marketing copy. GitHub runs whatever PR checks you have configured.
needsAn audit run and the GitHub integration for the repository that serves the site.
required inputsaudit_run_id, audit_revision, expected_repository, repository_serves_site
optional inputsfinding_ids, decisions, context

What the workflow does.The same steps, every run.

A Tin workflow is a repeatable job: declared inputs, a bounded run and a defined output.

  1. 01Re-reads robots.txt, the sitemaps and up to 40 pages, and drops findings that are already fixed.
  2. 02Asks your coding agent the judgment calls first, such as which URL survives a merge or whether a noindexed page should be indexed.
  3. 03Traces each finding to its source in your site's own framework, fixes it everywhere it applies and runs the repository's checks.
  4. 04Opens one pull request with the fixes by group, and lists copy changes and manual steps it left out.
  5. 05After you deploy, a later run checks each finding on the live site again.
ask your agent

Fix what our latest Tin audit found. Answer the judgment calls from the codebase and ask me only the ones you are unsure of.

What every workflowcomes 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

The judgment calls are where generic SEO automation goes wrong. Your agent answers them from the codebase and what it knows about the product, and asks you only the ones it cannot.

All workflows →

Using something else for this?What it replaces, and what it does not.

Questions.

Which findings can it fix?

Indexing directives, the sitemap, robots.txt, redirects and merges, titles, H1s and language tags from existing text, accessibility, structured data, Open Graph tags and internal links. Title and description rewrites go to the content workflows, and server or CDN problems are listed as manual steps.

Will it change my live site?

No. It opens a pull request. You merge and deploy on your own terms.

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.

Add Tin to your coding agent.It takes one command.

Try in browser →
codex mcp add tin --url https://app.tin.computer/mcp