Join the waitlist for the next waveThis wave of the closed beta is full. Drop your email and you're first in line when the next one opens.Enter your emailJoin the waitlist✓ You're on the list! We'll be in touch soon.Please complete the security checkYou're on the list.You're #{position} in line.We'll email you the moment the next wave opens — keep an eye on your inbox.Beta's full — join the waitlistJoin the waitlistA small closed beta — early access is limited.{filled} of {cap} beta spots filled· {waiting} waiting
How it works

Your automations, tailor-made

Your repetitive busywork handles itself. Describe it once in plain language, and Altmind takes it from there, on schedule or at the right moment.

No code requiredEvery run loggedYour model, your tools
altmind · new skill
# you write, in plain language:
"Every weekday at 7:30, read my inbox and
calendar, draft a brief, send it to me on
Telegram."
 
# altmind compiles it into:
trigger  → cron · 07:30 Mon–Fri
steps    → read(gmail, calendar)
         → draft(model: byom)
output  → send(telegram)
 
saved as daily-brief · live
The two pieces

Skill vs. automation: what's the difference?

They're easy to mix up. A skill is what to do. An automation is when to do it, on its own.

skill

A reusable capability

A task you describe once, in plain language: "summarize this thread", "draft a reply in my voice." Run it anytime, by hand, from chat, or via the API.

  • Written in plain language, versioned automatically.
  • Cheaper to run: no re-explaining means fewer tokens each time.
  • Uses your connected tools and shared memory.
automation

A skill that runs on its own

Add a trigger to a skill, and it fires without you, on a schedule, an event, a message, or an API call.

  • One trigger turns a skill into an automation.
  • Runs in the background, on Altmind's schedule.
  • Pause, edit or replay any run, anytime.
Anatomy of an automation

Trigger, steps, output: that's the whole shape

Every automation reads top to bottom. Something kicks it off, a few steps run in order with shared context, and a result lands wherever you asked. No diagrams to draw, no glue to wire.

  • Trigger, the "when": schedule, event, message or API.
  • Steps, the "do": read, think, act, using your tools.
  • Output, the "where": a message, a doc, a stored note.
Weekday 07:30
Cron trigger
when
Read inbox + calendar
Gmail · Google Calendar
do
Draft brief, flag replies
model: byom · shared memory
think
Send to Telegram
delivered 07:31
out
The building blocks

Four things every skill draws on

You don't configure these by hand. You mention them in plain language and Altmind wires them up. Knowing the four makes it easy to describe exactly what you want.

triggers

When it runs

A schedule (cron), an inbound message, a webhook event, or a direct API call. One skill can have several ways in.

schedule · event · message · api
tools

What it can touch

Your connected apps: inbox, calendar, Sheets, Docs, Telegram and more. Skills act through the tools you've granted, nothing else.

gmail · calendar · sheets · docs
memory

What it remembers

Shared memory carries context between steps and across runs: your preferences, past decisions, the people and projects you work with.

preferences · history · context
model

What does the thinking

Choose a model per skill: a cheap one for triage, a frontier one for drafting. Bring your own key and there's no token ceiling.

byom · per-skill · no cap
Reach

If it speaks API or MCP, Altmind can automate it

Altmind acts through tools. Any service that exposes an API or an MCP server can be wired into a skill: your inbox, calendar, CRM, docs, databases or internal systems. If you can reach it programmatically, Altmind can automate it.

What you can wire in
  • Email, calendar, chat: Gmail, Telegram, and the apps you live in.
  • CRMs, Sheets, Docs and storage: read, write, file away.
  • Databases and your own internal APIs.
  • Anything exposed over an MCP server, instantly in reach.
The honest caveat

Some platforms gate their own API.

Social media is the classic case. Posting to Instagram or Facebook isn't as simple as "call the API." Meta only allows publishing from Professional accounts linked to a Facebook Page, through an app that has passed Meta's review with the right permissions. Personal accounts can't post at all. That's Meta's design, not a limit in Altmind.

The fix is the one every social tool uses: connect a compliant publishing service (a social-media scheduler like Upload-Post) that has already cleared Meta's requirements and exposes its own API. Altmind drives that tool, and your post goes out.

Rule of thumb: when a platform restricts automation, the limit lives in the platform, and the way through is a tool that bridges it.

Build one

Three sentences from idea to live automation

There's no builder to learn. You describe the outcome, confirm the trigger, and turn it on. Altmind handles compiling, scheduling and running it, and shows you each run so you can tune the wording.

  • Edit the plain-language description anytime; it recompiles.
  • Test on demand before you schedule it.
  • Roll back to an earlier version in one click.
1 · Describe the outcome
"summarize my morning and message me"
say
2 · Confirm the trigger
weekdays · 07:30
set
3 · Turn it on
live · runs on its own
go
Trust & control

Nothing runs in the dark

Every run is recorded with its inputs, outputs and the steps it took. Inspect what happened, replay it with a tweak, or pause an automation entirely. Set a budget per skill so costs never surprise you.

  • Full run log: inputs, outputs, tokens, timing.
  • Replay any run, or pause an automation in one click.
  • Per-skill budgets and cost across providers.
Recent runs
daily-brief07:31 · 1.2k tok
inbox-triage09:14 · drafted 3
weekly-reportnext Fri 17:00
capturepaused
Starter automations

A few people set up on day one

Each is a single plain-language skill with a trigger attached. Copy the idea, change the words, make it yours.

daily-brief

Morning command center

Summarize inbox + calendar, flag what needs a reply, and push one message before you wake up.

07:30 → read → draft → Telegram
inbox-triage

Always-on triage

Every inbound email gets labeled, prioritized, and, if routine, auto-drafted in your voice.

on email → classify → draft
weekly-report

Reporting on autopilot

Pull numbers from Sheets, write the narrative, drop a ready-to-share doc in Drive.

Fri 17:00 → Sheets → Docs
capture

Catch every thought

Forward a voice note, photo or link from anywhere; it's transcribed, tagged and filed to memory.

on message → extract → store
follow-up

Never drop a thread

Spot promises you made in email and nudge you (or draft the follow-up) before they slip.

daily → scan → remind
research

Background research

Hand it a question; it gathers, cross-checks and returns a sourced brief to your vault.

ask → gather → brief
FAQ

Common questions

Do I need to write any code?

No. You describe what you want in plain language and Altmind compiles it into a running skill. Power users can still drop into the API and per-step settings for fine control when they want it.

What's the difference between a skill and an automation, really?

A skill is the capability: the thing to do. An automation is a skill with a trigger attached, so it runs on its own. The same skill can be run by hand and scheduled as an automation at the same time.

What can trigger an automation?

A schedule (like a cron time), an inbound message, a webhook event, or a direct API call. A single skill can be set up with more than one trigger.

Can I see and undo what an automation did?

Yes. Every run is logged with its inputs, outputs and steps. You can inspect it, replay it with a change, pause the automation, or roll the skill back to an earlier version.

Which model runs my skills?

Whichever you choose, per skill. Connect your own key for OpenAI, Anthropic, Gemini or a local model and run without a token ceiling. See For power users for the full picture.

Describe your first automation

One sentence in. A skill that runs itself out. Check the log whenever you're curious.