Outboundrix
Start free

AI on your key · no token markup

T-Ex is an operator,not a chat widget.

It drafts sequences from your ICP, writes first emails, classifies every reply and personalizes each send, on the same records your team works and with the model key you bring. It never skips a plan gate, a usage cap or the suppression list.

Start freeNo credit card required
Where MCP fits
T-Exrun · on your key
Series A foundersdraft · 4 steps
  • Day 0Quick question, {{first_name}}
  • Day 3One system instead of five
  • Day 8How {{company}} could cut tool spend
  • Day 12Should I close the loop?
Classify the replies that came in overnight.
classify_reply(× 12)3 meeting_request · 4 positive · 2 objection · 2 neutral · 1 unsubscribe
Each reply stopped that person’s sequence. The 3 meeting requests are at the top of Inbox.
personalize(ada@analyticalengines.io)“Saw Analytical Engines opened a London office…”
Runs onClaudeGPTGeminiGrokLocal
Illustrative run with example data.

Illustrative run with example data. Every call is a task T-Ex runs today, on your workspace’s own key.

What runs today

Four tasks, each where you already work.

No separate AI screen to learn. Each task sits on the page it serves and reads the same workspace data as the rest of the product.

Drafts a whole sequence

Your ICP and value proposition in; three to six steps with delays, subject lines and merge tags out, ready to edit.

sequence_from_icpSequences → Generate with AI

Writes the first email

One line about the goal becomes a sequence name, a subject and a first message.

sequence_starterSequences → Draft the first email

Classifies every reply

Positive, neutral, objection, unsubscribe or meeting request, on every reply that lands in the unified inbox.

classify_replyInbox

Personalizes each send

A per-contact rewrite at send time, cached per enrollment. If the model is unavailable or the cap is reached, the template goes out unchanged.

personalizeSending
Live todayPlanned with MCP
Bring your own model

Models the workspace can attach

One key per workspace, set in Settings → AI & T-Ex. Switch providers without touching a sequence.

Claude

Anthropic Messages API. Paste the key in workspace settings.

ChatGPT

OpenAI-compatible /v1. Your OpenAI or Azure base URL.

Gemini

OpenAI-compatible proxy or native when the workspace sets the URL.

Grok

xAI at https://api.x.ai/v1. Same OpenAI-compatible path. Your XAI_API_KEY stays in the workspace.

Local model

vLLM, Ollama or any OpenAI-compatible host. SSRF-guarded base URL.

MCP server

MCP is next. The scopes already exist.

The public REST API already scopes every key to contacts, sequences, campaigns, mailboxes and status. The planned MCP server exposes the same scoped operations to Claude, ChatGPT, Gemini, Grok or a local model, under the same API keys and permission model.

Until it ships, call the REST API with a scoped key or subscribe to HMAC-signed webhooks.

PlannedNot callable over MCP yet. Building sequences and campaigns, and the reply inbox, are dashboard-only today.
Read the API reference
Guardrails

What keeps the model in its lane

Each of these is enforced in the product, not promised in a policy.

  • Your key, encrypted

    Stored with AES-256-GCM and only ever shown back masked. It is used for this workspace’s calls and nothing else.

  • No token markup

    You pay your model provider directly. Outboundrix does not resell tokens.

  • Plan gates and a monthly cap

    Sequence drafting and send-time personalization need the Professional plan and count against the workspace’s AI-run cap.

  • Rate limited

    Every AI endpoint shares a per-workspace rate limit, so a script cannot run up your bill.

  • Suppression always wins

    The sender checks the suppression list before every send, whether a person or the model wrote the email.

API referenceTerms

Run the motion, not the stack.

Live scrape against your intent via Trusted Leads, then send from mailboxes you own. Free tier at launch — no card, no sales call.

Start freeNo credit card required