Mimir v0.4.0

0.3.0 was never published on its own, so updating from 0.2.0 also brings
everything in 0.3.0 (Signal, replies that survive
a restart and more).

Talking to your bots

• Telegram rich messages: replies show real tables, headings and task lists, and a long one fits in one message
  (up to 32,768 characters). An older Bot API server gets the formatting it had before.
• Slack streams natively: in a thread you mentioned the bot in, a reply streams in with Slack's own streaming
  (no edit rate limits) under an "is thinking…" status.
• Skills for your bots: four more in the skill library. Hire a bot makes a bot with one job, what it never does,
  a voice and when it works, and has it say its job back; Bot role check checks a role and rewrites it; Team check
  finds where your bots overlap or stray and proposes fixes you pick; Memory check proposes fixes for wrong, stale or
  secret memories.

Routines

• Goals: ask a bot to see something through ("find a plumber who can come this week") and it keeps at it until
  it's done, with a progress line pinned in the chat and in its header. The follow-ups it books and the tasks it hands to
  other bots belong to the goal: pause it and they all stop at once, even mid-task; remove it and
  they're gone. A bot can only call a goal done when the run log shows it, and a goal can have a token budget of its own.
  When the budget runs out, Needs you asks whether to go on. More
• Routines remember their last results: each run sees what the routine said its last three times, so a daily
  briefing says what's new instead of repeating yesterday's news.

Safety

• Sensitive sources: mark a tool server (your bank's, your health records'), a single tool or a chat's memory as
  sensitive in Settings → Sensitive sources. Once a reply or routine has read from one, everything that goes out
  asks first, and the card says why ("This run read from bank"): an email, a web page or search, a task for another
  bot, a message to someone else's chat, and any tool server's change. Servers whose names look like banking or
  health are suggested, never marked for you.
• Security: the TLS library (rustls 0.23.46) is updated for four advisories published on October 14
  (RUSTSEC-2026-0346 to 0349).
• Edit an email before it goes, and say why when you deny: on an email's approval card in the app, Edit lets
  you change the subject and text before you send it. What you approve is what goes; who it's to stays as shown. In
  Telegram, Slack, Discord and Signal, Edit in Mimir opens the card in the web app (on a server with an https
  address). Deny… takes an optional "Why?", and in a messaging app a reply to the card does the same: the bot
  reads your reason and changes course instead of just stopping.
• Export, import and duplicate a bot: right-click a bot to save it as one file (its role, look, skills, routines,
  tool rules and the tool servers it uses), without tokens, keys, saved logins, memories or chats. From a file… on
  the new-bot page shows everything in it first, and flags skill steps that pipe a download into a shell, hold a base64
  blob or download a program. Skills come in as proposals and routines come in paused; only its tool rules that are at
  least as strict as yours can be used, and its AI account uses your own settings.
  Duplicate… does both in one go. See Bots and chats.
• Grok bots are more private: Grok's own memory and /goal are always off for bots, and your Claude and Cursor
  settings no longer load into them, so bots can't read each other's notes and incognito chats leave nothing behind
  in Grok. A message that starts with / now reaches the bot instead of running as a Grok command. If a Grok update
  ever offers a bot more tools than web search and Mimir's own, the message stops with an error instead of running.
  The docs cover The X Subscription's plans and its shared weekly allowance.
• Security fixes from the fourth review:
  • A routine can no longer change its own times or end date while it runs, or book itself again a minute later;
    from a routine, a one-off is at least 15 minutes away.
  • A follow-up stays in the chat it was booked in, and a reminder for someone else counts toward their chat's limit.
  • With docker compose, Codex bots now use the separate computer service; before, they started a browser inside
    Mimir's own container.
  • After reading a sensitive source, saving a memory for the household asks too.
  • A bot file can't pass an option to the AI's command line as its model.
  • mimir doctor no longer creates a database, or upgrades an older one, while it checks.
• Anonymous usage stats: once a day Mimir sends a count (a random id for this install, the version, the OS and each
  bot's AI provider and model; never names, messages or keys) so we know how many people use it and on what.
  Settings → Your data shows exactly what's sent and switches it off (MIMIR_USAGE_STATS=off on a server).
• Security checks in mimir doctor: it warns when the web app is reached over plain HTTP from the internet,
  Permissions are at "Allow all", a tool server's package has no fixed version, a newer Mimir is out, or the bots'
  computer runs in Mimir's own process on a Linux host without Docker (its DevTools have no token there).
• Tool servers stay on one version: adding one that downloads its package (npx -y pkg, uvx pkg) pins it to
  the newest release that's at least a week old, so a bad release published tomorrow doesn't reach you on the next
  start. When a newer one has been out a week, the server shows Update to x.y.z.
• Hidden text doesn't reach your bots: invisible characters that spell out words only a model reads (Unicode tag
  characters, stray variation selectors) are taken out of pages, mail, files and tool results. A Word file's text
  formatted to be invisible (white, 1 pt, hidden) is marked as hidden for the bot.
• A tool server's "read-only" needs your OK: a server saying a tool only reads is no longer enough to skip asking;
  after you add a server, Settings → Tool servers asks you to confirm which of its tools really only read.
• More security fixes:
  • Signing in to a tool server checks the answer comes from the server you started with (RFC 9207).
  • Codex's own web search answers from OpenAI's cache only, and is off after a bot has read outside content, so it
    can't open an address that carries your data out.
  • Email approval cards flag addresses with look-alike letters from another alphabet and show their real domain.
  • Only the chat's owner can answer a bot's buttons, and screenshots are kept in Mimir's own folder instead of /tmp.

The bots' computer

• Sign in from your phone: at a sign-in page, a bot sends you a card with a link to a small masked form in Mimir,
  instead of asking you to take over the screen. Type your login there, never in the chat; Mimir types it into the
  page and keeps it for that bot on that site only. The bot never sees it, only whether it worked. If the bot got to
  that site after reading a web page or a message, the site looks like one you've signed in to before, or its
  address uses letters from another alphabet, the form warns you and shows the full address. Taking over stays for
  captchas and payments. Details
• Authenticator codes on saved logins: paste a site's setup key or otpauth:// link once, and Mimir types the
  6-digit codes itself, only on that login's site.

---

Installers for macOS (Apple silicon), Windows and Linux. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

Mimir v0.2.0

Talking to your bots

• Change course mid-reply: send "actually, only this week" while a bot is still answering and the reply in progress
  takes it in, instead of waiting until it's done. The chat shows Changing course… until the new answer starts.
• Tap instead of type: in Telegram, Slack and Discord a bot can ask with 2 to 4 buttons (plus Something else),
  and routine and reminder results have one-tap feedback buttons.
• Bots send files and screenshots back: a file from the chat's workspace, or a screenshot of its computer's screen
  after a booking or a filled-in form, comes with the reply.
• Receipts under replies: in a messaging chat, each thing that really happened gets a line ("Sent an email to …",
  "Reminder set for …") and the pages and searches it used. A reply that says it sent or booked something it didn't
  gets a note: Not done in this reply.
• What a reply used: under each reply in the app, "2 memories · 1 skill · read 3 pages". Open it to edit or forget
  a memory that steered it wrong. In a messaging app, /why lists the same.
• A busy bot doesn't leave other chats in silence: it shows it's typing everywhere at once and works on each chat
  in turn.
• A fresh start after a quiet spell: a Telegram, Slack or Discord chat that's been quiet for 12 hours starts a new
  session, opening with where you left off.
• Telegram topics: turn on Each topic is its own conversation on a bot's page, and every topic in your private
  chat with it gets its own conversation.
• Slack threads say which message: a reply in a thread reaches the bot with the message that started it.
• Mimir in Dutch: Settings → Language. Pairing, approval cards, errors, the digest and the Telegram menu follow it.
  Bots already answered in your language.
• Chats show earlier messages, and you can search them.
• On a phone: a tab bar at the bottom (Chats, Needs you, Memory, Routines) with the sidebar as the Chats screen.
  Before, the web UI on a phone had no way to get to your bots.
• Push-to-talk: in the chat or quick ask, hold ⌥ (Ctrl on Windows and Linux) and speak; let go and the words are
  in the box, not sent. Transcribed on your computer by the same whisper.cpp as meetings.
• Voice notes without an API key: with no OpenAI key, Telegram and Discord voice notes are transcribed by
  whisper.cpp on your computer.
• Word and Excel files and scanned PDFs: send a .docx or .xlsx and the bot reads it (Excel sheet by sheet).
  A scanned PDF's pages go to the bot as photos, so a scanned contract or letter no longer needs photographing.

Memory

• Memories for one bot only: "just for you: …", or switch a memory between all bots and only on its card.
• The Household: shared memories and lists (shopping, plans, who does what) your bots read and add to, and family
  group chats that use them.
• Pinned cards: a bot keeps a card at the top of a chat up to date, like a shopping list or bills due. "Add oat milk
  to my shopping list" rewrites the card instead of sending a message.

Routines

• Started by email or a webhook: a routine's schedule can be mail:from:acme.com subject:invoice, so it runs when
  matching mail arrives (mail that doesn't match costs no AI), or webhook on your server (GitHub-compatible
  signatures). "Tell me when the invoice from Acme comes in" sets one up.
• Timed off your calendar: calendar:60 runs an hour before each event; calendar:60:meetings only when someone
  else is invited, calendar:60:outside only when someone from outside is. "Brief me on who they are before each
  outside meeting" now works.
• Urgent results come through right away, on your rule: a routine that waits for the digest or quiet hours can carry
  one line, "Come through right away if my flight is delayed". At most 3 a day.
• Approvals on your phone: Settings → Decisions sends every routine approval to one private chat as well, so one
  from the app chat doesn't wait until you open the app.
• A fallback account at the usage limit: on a bot's page, under When it hits a usage limit, routines move to
  another account while the bot's own is at its limit, and come back when it resets.
• Pause: all bots (next to Bots in the sidebar) or one (its right-click menu), until you resume, for an hour,
  until tomorrow or until Monday. Nothing runs on its own meanwhile; you can still ask them directly.

Email, tools and your server

• Tasks by email: forward or CC mail to you+mimir@…, or file it in a Mimir folder, and your note on top is
  the task.
• More than one mailbox, and Outlook (with your own Microsoft sign-in).
• Tool servers that need a sign-in: hosted MCP servers such as Notion and Linear sign in from Settings.
• A chat API: mimir server answers OpenAI-style /v1/chat/completions for Shortcuts, Home Assistant and the like,
  with a token per client that reaches one bot (Settings → Chat API).
• Backups you can restore: Mimir checks its database every day and keeps 7 daily and 4 weekly copies. Settings →
  Your data → Restore a backup…; damage shows in Needs you.

First run

• A starter team in the first minutes: in the first minutes:** the first run asks what you want off your plate (tap up to three), and your
  first bot suggests bots, routines and library skills for them on one card. Untick what you don't want, and the rest
  is set up when you start.
• Get to know me (Settings, once email or the calendar link is connected): one read-only look at who wrote to you
  this month and what's on your calendar. What it finds waits in Memory as suggestions, and nothing is kept until you
  accept it.
• Bots know what isn't connected yet: ask for your email before it's set up and the bot says where to connect it.
  The chat shows a button to that setting and a Try again for afterwards.
• Email providers that send on port 587 work: Settings → Email has an Outgoing port next to the outgoing server.
  Empty is 465 as before; 587 (or any other port) uses STARTTLS. Saving used to set it back to 465 every time.

Faster and steadier

• Faster replies: the tools process starts in about 25 ms instead of 1.3 s, and the app launches faster.
• The app asks less: it polls only the view you're looking at, one request at a time, and not at all while hidden.
  The web UI is compressed and cached.
• A crash in one routine fails that run, not the whole bot, and a bot that crashes starts again by itself.
• A chat past 200 messages no longer goes blank.
• A database from a newer Mimir is refused with a clear message instead of being half-upgraded.
• The quick-ask panel sees bots added, renamed or deleted since it opened, and its bot can't be changed mid-answer.
• Guest/Yours on a chat and "Open in Mimir" say so when they fail; a second "are you sure?" no longer leaves the first
  one hanging.

---

Installers for macOS (Apple silicon), Windows and Linux. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

Mimir v0.1.5

Installers for macOS (Apple silicon), Windows and Linux. The Mac app is signed and notarized by Apple. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

Mimir v0.1.4

Installers for macOS (Apple silicon), Windows and Linux. The Mac app is signed and notarized by Apple. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

New in 0.1.4: Accept all / Allow all for memory suggestions, memories to confirm, approvals and permissions; a self-hosted server tells you when to update it. Details: https://github.com/currentbits/mimir/blob/master/CHANGELOG.md

Mimir v0.1.3

Installers for macOS (Apple silicon), Windows and Linux. The Mac app is signed and notarized by Apple. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

New in 0.1.3: Export and Copy work in the desktop app; switching chats or bots never shows the one before; buttons that fail say why. Details: https://github.com/currentbits/mimir/blob/master/CHANGELOG.md

Mimir v0.1.2

Installers for macOS (Apple silicon), Windows and Linux. The Mac app is signed and notarized by Apple. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

New in 0.1.2: deleting routines, bots, files, meetings and more works in the desktop app again (the "are you sure?" now asks inside the app). Details: https://github.com/currentbits/mimir/blob/master/CHANGELOG.md

Mimir v0.1.1

Installers for macOS (Apple silicon), Windows and Linux. The Mac app is signed and notarized by Apple. The Windows installer isn't code-signed yet: choose More info, then Run anyway the first time. How to use Mimir: https://heymimir.app/docs/

New in 0.1.1: the Mac app is notarized by Apple; search in Settings; each bot opens its own conversation. Details: https://github.com/currentbits/mimir/blob/master/CHANGELOG.md

Mimir v0.1.0

Installers for macOS (Apple silicon), Windows and Linux. They aren't code-signed yet: on a Mac, right-click the app and choose Open the first time; on Windows, choose More info, then Run anyway. How to use Mimir: https://heymimir.app/docs/

The first public release. What it does: https://heymimir.app · what is in it: https://github.com/currentbits/mimir/blob/master/CHANGELOG.md