Solving tickets with Claude AI

Claude reads the conversation, checks your code, and drafts the reply in your voice. You say send.

Works with Claude, on the Pro or Max plan you already have.

80 seconds. A real GoodSign customer question, answered from Claude Code.

What just happened

Nothing staged: a real customer, GoodSign’s real code, and the reply went out.

  1. The question

    Mark asks how to pre-fill a field through the API and lock it. Not a FAQ answer.

  2. One line

    “Pull down the latest issue from Mark Silcock.” Claude gets the whole thread.

  3. Checked against the code

    Claude reads the actual handler. The right answer, plus one gotcha the docs miss.

  4. Send it

    A draft in John’s voice. “Send it.” Mark has his answer.

Bug reports, fixed before they leave the page

A visitor reports a bug with a screenshot. get_conversation brings the report, the pages they were on and when, their browser and OS, and the screenshot itself into your terminal.

Claude Code is already in your source with your logs a command away, so it goes from “the button does nothing” to the line that broke it, fixes it, and drafts the reply. The customer hears back before you’ve switched windows.

claude code

Answer from wherever Claude is

The same connection works in the Claude app on your phone, on claude.ai and on the desktop. Ask what came in, read a thread, get a draft in your voice, say send. tiny.help doesn’t even need to be open.

Set your voice notes once and every draft reads like you wrote it, whichever device you’re holding.

Claude app

The tools. One rule.

tiny.help is a remote MCP server. Claude already knows how to use it; there’s nothing to install on your side beyond the connection.

  • list_conversationsYour inbox, newest first, with a snippet of the last message. Filter by site, by open or closed, or by one customer’s name or email.
  • get_conversationOne full thread with the quoted-email pyramids already stripped, the visitor’s browser, OS, location and recent pageviews, and attachment URLs Claude can open to look at screenshots.
  • get_voice_notesYour tone, stock phrases, sign-offs and the things you never say. Claude is told to read this before drafting anything as you.
  • send_replySends an already-approved reply as you. The customer sees it in their chat widget immediately; if they’ve gone, the email pipeline follows up.
  • react_to_messagePuts a ❤️ 👍 👎 😂 or ❓ on one message, the way a tapback works in Messages. The customer sees it in their chat widget right away, and it can be taken back.
  • close_conversationWrap it up. A new message from the customer reopens it on its own.
  • list_shortcutsThe saved replies on each of your sites: the stock answers your team expands in the composer by typing /pricing.
  • save_shortcutWrite a new saved reply, or rewrite one you already have. Claude hands back the wording it replaced, so an overwrite is never a one-way door.
  • delete_shortcutRetire one. It comes back with the body it removed, in case you want it again.

And the setup tools, for the day you connect: say "Set up tiny.help for https://yoursite.com" and Claude walks you through it.

  • setup_statusWhere the account stands and what is left to do, with the plans when none is started.
  • start_checkoutThe Stripe link for a 14-day trial. You add the card; it comes back to Claude.
  • inspect_siteReads your homepage: brand colours, sitemap shape, sign-in links, whether the widget is already there.
  • create_site, update_site, get_site, list_themesThe site, its welcome message, colours, quick links, live monitoring and the embed snippet.
  • update_knowledge, start_crawl, list_pages, delete_pagesWhat to index and what to skip, the crawl, and pruning what got in.
  • get_faq_guide, upload_faq_filesA prompt for writing your support guides from your code or your docs, shown to you in full first, and the upload once you have read every word of the result.
  • invite_member, set_member_sites, list_teamYour team, inside the plan's limits.
  • set_assistantSwitches the assistant on for a site, once there is something for it to answer from. It is off until you ask for it.

You approve every send. That’s not a marketing line; it’s the instruction the server gives Claude with the send tool, every time:

“This delivers to a real person and cannot be unsent. Never use this tool to compose, preview, or ‘show’ a reply. Write drafts in the chat as plain text instead. ‘Reply to #3’, ‘draft one’, silence, or your own suggestion to send are none of them approval. If you are unsure whether you have it, you do not. Draft and ask instead.”
This is the rule for Claude writing as you: a person reads the draft, and a person says send. The assistant is the other thing, and it is deliberately not the same. It answers on its own, but only from pages you approved, only in four categories, and it goes quiet the second one of your people types. The one with your voice never sends without you; the one that talks to customers can only ever quote you.

Connect once

Two ways in, depending on where you want your desk. Both use the same sign-in and the same access rules as the desk itself.

Claude Code

One computer, one command. Adds nothing to your Claude account or phone.

claude mcp add --transport http \
  tinyhelp https://desk.tiny.help/mcp
then /mcp → authenticate

Sign in with the Google account you use for your desk. From then on Claude Code can work a ticket while it’s sitting in your project’s code.

Phone, claude.ai and the desktop app

Added once in a browser, it’s on your whole Claude account.

  1. Open claude.ai → Settings → Connectors (Pro or Max plan).
  2. Add custom connector → paste the URL above.
  3. Sign in if asked, with the Google account you use for your desk. Whoever presses Allow is who Claude acts as.
  4. Press Allow. On your phone it now appears in a new chat under the + button → tools.

Questions

Do I need a Claude subscription?

No. The assistant that answers your customers runs on our account and is included on every plan. A Claude Pro or Max plan is what lets Claude do the setup: read your code, write your support guides, and upload them once you have read them. It also lets you work a thread from your terminal. Both are worth having and neither is required.

Does Claude ever send a reply on its own?

No. The send tool only accepts a reply you’ve already approved, and its own instructions tell Claude that a draft, “reply to #3”, or silence isn’t approval. If Claude is unsure whether it has approval, it doesn’t. It drafts and asks. The assistant is separate and does answer on its own, from your indexed pages only, and only where you have switched it on.

Does tiny.help read my source code?

No. tiny.help gives Claude the conversation: the messages, the visitor’s browser and pages, any screenshots. Claude Code reads your code because it’s running in your repository. That’s what makes the answer right.

Is tiny.help an MCP server?

Yes. A remote MCP server over HTTP at https://desk.tiny.help/mcp, with OAuth sign-in and the tools above. It’s built for Claude Code, the Claude apps and claude.ai.

Can my teammates use it too?

Everyone on your desk connects with their own sign-in. Drafts come out in their voice notes, and a send is theirs. The customer sees who replied.

What does the connector cost?

It’s included on every plan. $29 a month for three sites and three people, with a 14-day trial. Drafting runs on your Claude plan.

Try it on your own inbox

Install the widget, connect Claude, and answer the next hard question from wherever you are. 14 days free, card up front, nothing charged until day 15.

Start your free 14-day trial