Skip to content

TalkwardTalkward for MCP

Your inbox for every AI agent

The Talkward MCP server gives AI agents access to your inbox through the open Model Context Protocol. Claude Code, Claude Desktop, Cowork, ChatGPT or your own agent search conversations, read histories, write notes and set tags. With the “Set up” right your AI also sets Talkward up and moves past conversations over from your previous system. Every agent with its own key, revocable at any time.

HaferhainAI assistant replies instantly
WhatsApp

Hi, my muesli subscription arrived twice this month. Can I send one back?

Claude Code · key “Support agent”

Double delivery caused by two active subscriptions. Suggestion: end the second one, free return.

Tag “Double subscription” · set by Claude Code

Mira

A second subscription slipped through, sorry! I have ended it, the return label follows in a moment.

Write a message

Sound familiar? Your AI agent can do a lot, except look into your inbox.

You work with Claude or ChatGPT every day. But for every customer service question you copy conversations back and forth.

  1. Copy, paste, ask.

    To summarise ten complaints you paste ten transcripts into the chat. Then you copy the answer back.

  2. A custom integration for every agent.

    Your developer builds an agent that pre-sorts requests. First he writes an API client, with sign-in, paging and error cases.

  3. Nobody sees what the AI did.

    A script sets tags and writes notes. Who did it and when is recorded nowhere.

With the MCP server you connect an agent in a minute. Everything it changes shows in the conversation with its name.

Tools: 26 tools for your agent.

These are all the tools of the server. Reading tools work with any key, writing ones only if the key allows it. The second group needs the “Set up” right, which only the owner grants.

Inbox: read, note, reply

ToolWhat it doesKind
gespraeche_suchenSearch conversationsreads
gespraech_lesenRead a conversation with its historyreads
kontakte_suchenSearch contactsreads
kontakt_lesenRead a contactreads
auftraege_offenFetch open AI jobsreads
notiz_schreibenWrite an internal notewrites
antwort_sendenSend a reply to the customer, not with “read only”writes
status_setzenSet statuswrites
tags_setzenSet tagswrites
auftrag_abschliessenComplete a job, report the resultwrites

Set up and move: plan first, then apply, undoable at any time

ToolWhat it doesKind
arbeitsbereich_standState of the workspace: plan, limits, brands, team, channels, agents, automationsreads
marke_anlegenCreate a brandwrites
marke_aendernChange or archive a brandwrites
mitarbeiter_anlegenInvite a team member, with role and brandswrites
abteilung_anlegenCreate a departmentwrites
kanal_anlegenConnect an email mailbox or WhatsApp numberwrites
ki_agent_anlegenCreate an AI agent with knowledge basewrites
ki_agent_aendernChange an AI agent, switch it on or offwrites
textbausteine_setzenSet the snippets of a brandwrites
kontakte_anlegenCreate contacts, up to 500 per callwrites
gespraeche_importierenImport past conversations with messages and attachmentswrites
automation_entwerfenHave an automation drafted from a descriptionwrites
automation_anlegenCreate an automationwrites
automation_schaltenSwitch an automation on or offwrites
einrichtung_protokollLog: what the AI has set upreads
einrichtung_rueckgaengigUndo a stepwrites

Set up: One line or one file.

In Claude (web, desktop, Cowork) Talkward is in the connector directory: search for “Talkward”, connect, sign in with Talkward. In ChatGPT you enter the URL as a connector. In Claude Code a command or a file in the project with a key is enough.

Claude Code, by command
claude mcp add --transport http talkward https://chat.talkward.app/api/mcp \
  --header "Authorization: Bearer tw_…"
As .mcp.json in the project
{
  "mcpServers": {
    "talkward": {
      "type": "http",
      "url": "https://chat.talkward.app/api/mcp",
      "headers": { "Authorization": "Bearer tw_…" }
    }
  }
}

What agents do with it

Five ways to work Talkward with AI.

The same server for questions in chat, jobs from automations, your own agents and the move.

  1. Questions to your inbox

    In Claude or ChatGPT: “Summarise all open complaints from today.” The agent searches and reads on its own.

  2. Drafts for approval

    “Reply to the three oldest conversations in our brand voice, as a note for approval.” Nothing is sent until your team wants it.

  3. Jobs from automations

    The block “Claude Code beauftragen” creates a job. The agent reads the conversation, works and reports back with auftrag_abschliessen.

  4. Your own agents

    Your agent pre-sorts, sets status and tags or prepares replies, without its own API client.

  5. Move and set-up

    “Read my staff list and invite everyone” or “Fetch the last twelve months of tickets from my old helpdesk and bring them over with messages and attachments.” The AI shows the plan first, you confirm, and every step is listed in Talkward with “Undo”.

How it works: One server, one key per agent.

The MCP server runs at https://chat.talkward.app/api/mcp and speaks the Model Context Protocol over HTTP. An agent signs in with a Talkward API key and gets exactly what the key allows. Every change shows in the conversation with the key’s name.

  • One key per agent

    Every agent gets its own Talkward API key. Its name shows next to everything it does in the conversation.

  • “Read only”, “Notes only”, “Set up”

    “Read only” searches and reads. “Notes only” also writes notes, tags and results, but never to customers. “Set up” is granted by the owner only: with it the AI creates brands, team, channels, AI agents, snippets and automations and imports past conversations, always as a plan first.

  • Revocable at any time

    Revoke the key and the agent is out immediately. Everything the AI set up is listed under “Set up by the AI” and can be undone step by step.

  • Connected to automations

    The agent fetches jobs from “Claude Code beauftragen” with auftraege_offen and completes them with auftrag_abschliessen.

Connection
Model Context Protocol over HTTP at https://chat.talkward.app/api/mcp, bearer key per agent
Access
26 tools: search and read conversations and contacts, note, reply, status, tags, jobs; with “Set up” brands, team, channels, AI agents, snippets, automations and migration of past conversations
Processing
Talkward: Nuremberg (ISO 27001). What an agent reads is processed by its provider; for Claude and ChatGPT in the USA, under your own DPA.

What an agent leaves in the conversation.

Notes, status and tags carry the key’s name. Your team sees which agent did what and can take its access away at any time.

  • Key per agent
  • “Read only”
  • Name in the conversation
  • 26 tools
  • Revocable any time
  • Set-up and migration
CC
Claude CodeKey “Support agent”
MCP
Rights
Read and write
Note
Double delivery caused by two active subscriptions
Tag
Double subscription
Sender
Claude Code · Support agent
Revocation
any time, effective immediately

Set up in four steps.

The MCP server is included from Professional at no extra cost.

  1. Switch on the add-on

    Under Settings › Add-ons switch on the matching add-on: Claude, ChatGPT or the API for your own agents.

  2. Choose the rights

    With “Sign in with Talkward” or a key per agent: “Read only”, “Notes only”, “Read and write” or, as the owner, “Set up” for set-up and migration. Every connection can be disconnected under Add-ons.

  3. Add the connector

    In Claude Desktop, Cowork or ChatGPT as a connector with the URL https://chat.talkward.app/api/mcp, sign-in via OAuth. In Claude Code by command with a key as bearer token.

  4. Get going

    Ask the agent about your inbox or drag “Claude Code beauftragen” into an automation.

MCP server

no extra costfrom Professional, provisional, net

No extra cost from Professional. The models of your agents are billed by their providers through your account.

All plans and prices
  • 26 tools for conversations, contacts, jobs, set-up and migration
  • Model Context Protocol over HTTP
  • Key per agent, from “read only” to “set up”
  • Every change in the conversation with its sender, every set-up step undoable
  • Connected to “Claude Code beauftragen”
  • Servers in Nuremberg (ISO 27001)
  • Data processing agreement
  • AI replies labelled (EU AI Act)
  • Extensions cancellable monthly

Questions about the MCP server

Question not listed? Write to us at hey@talkward.app.

Which AI agents work?

Any that speaks the Model Context Protocol over HTTP: Claude Code, Claude Desktop and Cowork, ChatGPT with connectors in developer mode, and your own agents.

How does an agent sign in?

With a Talkward API key as a bearer token. Create one key per agent, then its name shows next to everything it does in the conversation.

Can I limit access?

Yes. A “read only” key searches and reads but sends no replies to customers. Every key can be revoked at any time, and the agent is out immediately.

What does the MCP server cost?

Nothing extra, included from Professional. Claude, ChatGPT or other models are billed by their provider through your own account.

Where is the data processed?

The MCP server is part of Talkward, storage stays in Nuremberg. What an agent reads is processed by its provider; Claude and ChatGPT are services in the USA with which you sign your own data processing agreement.

Can the AI set Talkward up for me or handle my migration?

Yes. With a “Set up” key, granted by the owner only, your AI creates brands, team members, departments, email and WhatsApp channels, AI agents, snippets, contacts and automations and imports past conversations from your previous helpdesk with messages, attachments, status and the original timestamps. Your AI reads the old system itself, for example through its API or an export. Every step comes as a plan you confirm first and is then listed in Talkward under “Set up by the AI” with “Undo”. Imported conversations trigger no automations, notifications or AI replies and do not count towards the contact quota.

What are auftraege_offen and auftrag_abschliessen for?

For jobs from automations: the block “Claude Code beauftragen” creates a job, the agent fetches open jobs and reports result and outcome with auftrag_abschliessen. The automation then continues.

Model Context Protocol is a trademark of its respective owner. Talkward is an independent product and is not operated, sponsored or endorsed by Model Context Protocol. Model Context Protocol and the MCP logo are trademarks of LF Projects, LLC. For more information, visit https://modelcontextprotocol.io

Secure your seat.

Three details, no commitment. We will get back to you personally within two working days.

hey@talkward.app · WhatsApp

What would you like?

We use your details only to reply to you. More in the privacy policy.