{
  "name": "Autark",
  "formerNames": [
    "Proactive Agent Workforce",
    "Workforce"
  ],
  "url": "https://metajarvis.teydralab.com/autark/",
  "descriptor": "Your own AI company, on your own computer",
  "summary": "Autark is a free desktop app that gives one person their own AI company: a manager, specialist agents and a private secretary that work proactively — they wake up on schedules, messages, tasks and events instead of waiting for a prompt. It runs on the user's own computer and uses the AI plans the user already pays for (ChatGPT plans through OpenAI Codex, paid Claude plans through Claude Code). Agents have full access to the PC: there is no permission gate, guardian model or money limit, and text steers them. Settings, skills and memory are open files the AI can change itself. Made by Teydra Lab (Salzburg, Austria); part of Meta Jarvis. Formerly called \"Proactive Agent Workforce\".",
  "maker": {
    "company": "Teydra Lab",
    "url": "https://teydralab.com/",
    "founder": "Stanislaus Hees",
    "founderUrl": "https://stanislaushees.com/",
    "location": "Salzburg, Austria",
    "umbrella": "Meta Jarvis",
    "umbrellaUrl": "https://metajarvis.teydralab.com/"
  },
  "price": {
    "app": "free",
    "account": "none",
    "aiUsage": "billed by your own AI providers"
  },
  "platforms": {
    "windows": {
      "version": "0.3.2",
      "date": "2026-11-05",
      "sizeMb": 132,
      "status": "current",
      "signed": false,
      "arch": "x64"
    },
    "macos": {
      "version": "0.3.2",
      "date": "2026-11-05",
      "sizeMb": 272,
      "status": "current",
      "signed": "ad hoc, not notarized",
      "arch": "universal",
      "computerControl": false,
      "selfUpdate": false
    },
    "linux": {
      "version": "0.3.2",
      "date": "2026-11-05",
      "sizeMb": 158,
      "status": "current",
      "arch": "x64"
    }
  },
  "downloads": {
    "windows": {
      "stable": "https://metajarvis.teydralab.com/autark/download/windows",
      "direct": "https://github.com/teydra/proactive-agent-workforce-releases/releases/latest/download/autark-windows-x64.exe"
    },
    "mac": {
      "stable": "https://metajarvis.teydralab.com/autark/download/mac",
      "direct": "https://github.com/teydra/proactive-agent-workforce-releases/releases/latest/download/autark-mac-universal.dmg"
    },
    "linux": {
      "stable": "https://metajarvis.teydralab.com/autark/download/linux",
      "direct": "https://github.com/teydra/proactive-agent-workforce-releases/releases/latest/download/autark-linux-x64.AppImage"
    },
    "checksums": {
      "stable": "https://metajarvis.teydralab.com/autark/download/checksums",
      "direct": "https://github.com/teydra/proactive-agent-workforce-releases/releases/latest/download/SHA256SUMS.txt"
    }
  },
  "engines": {
    "available": [
      "OpenAI Codex (ChatGPT plan)",
      "Claude Code (paid Claude plan)"
    ],
    "early": [
      "API keys: OpenAI, Anthropic, Google, OpenRouter, DeepInfra, Groq, Mistral, xAI, DeepSeek, Kimi, Qwen, any server speaking a supported protocol",
      "Local servers: Ollama, LM Studio, llama.cpp, vLLM",
      "Plan CLIs over ACP: Gemini CLI, GitHub Copilot CLI, Kimi Code (never seen in a live run)"
    ]
  },
  "access": {
    "sandbox": false,
    "approvalPrompts": false,
    "permissionGate": false,
    "guardianModel": false,
    "moneyLimit": false,
    "firstQuestion": "a question in the chat, not a technical barrier",
    "stopAllAgentWork": "this PC only, holds across a restart"
  },
  "privacy": {
    "dataLocation": "user's own computer",
    "vendorServers": false,
    "analytics": false,
    "appReadsProviderTokens": false
  },
  "openSource": false,
  "statusLabels": {
    "available": {
      "label": "Available",
      "long": "In the current public release"
    },
    "next": {
      "label": "Next release",
      "long": "Built in development, ships with the next release"
    },
    "planned": {
      "label": "Planned",
      "long": "Not built yet"
    }
  },
  "capabilities": [
    {
      "id": "team",
      "group": "Your AI company",
      "title": "Manager, teams and a secretary",
      "status": "available",
      "text": "Start with a manager. It builds the team you need (research, marketing, finance, a personal secretary) and hands work to the right person. Agents have names, avatars and their own instructions in a file."
    },
    {
      "id": "delegate",
      "group": "Your AI company",
      "title": "Delegation between agents",
      "status": "available",
      "text": "Agents pass tasks to each other, ask colleagues for help and report back when the work is done. You talk to one of them; the organisation does the routing."
    },
    {
      "id": "split",
      "group": "Your AI company",
      "title": "An agent can split itself",
      "status": "available",
      "text": "For work that runs well in parallel, an agent decides itself to split into parts, each working on its own, and brings the results back together."
    },
    {
      "id": "import",
      "group": "Your AI company",
      "title": "Bring agents you already have",
      "status": "available",
      "text": "Hand your manager Claude Code subagents, a Codex AGENTS.md, a CrewAI YAML file or a list of roles. A skill shows it how to bring them into the team."
    },
    {
      "id": "people",
      "group": "Your AI company",
      "title": "Humans and agents on one team",
      "status": "available",
      "text": "Invite colleagues by link or QR code. People are round, AI is square, and everyone works in the same chats and tasks. The data stays on your PC."
    },
    {
      "id": "chats",
      "group": "Chats and work",
      "title": "Chats and group chats",
      "status": "available",
      "text": "Every agent has a direct chat; add groups and as many chats per agent as you like. You can keep typing while agents work; a message marked important steers work that is already running."
    },
    {
      "id": "rooms",
      "group": "Chats and work",
      "title": "Three message levels",
      "status": "available",
      "text": "Messages can be normal, important or quiet, in both directions, and every reply shows which model wrote it."
    },
    {
      "id": "work-inbox",
      "group": "Chats and work",
      "title": "Work: one inbox for every task",
      "status": "available",
      "text": "Type or dictate. With Auto your manager takes it, decides who does it and can split it into separate tasks. Or pick an agent yourself, or Me for your own to-do list."
    },
    {
      "id": "my-day",
      "group": "Chats and work",
      "title": "My day",
      "status": "available",
      "text": "Your own to-dos and calendar in one calm list, with learned time estimates, a morning plan and a short briefing of what happened."
    },
    {
      "id": "team-view",
      "group": "Chats and work",
      "title": "Team",
      "status": "available",
      "text": "Every agent and person, what they are working on and what they did. An agent’s page opens its instructions and its live browser."
    },
    {
      "id": "calendar",
      "group": "Chats and work",
      "title": "Calendar",
      "status": "available",
      "text": "Read any ICS feed, or connect Google Calendar both ways so agents can plan around your real day."
    },
    {
      "id": "schedule",
      "group": "Chats and work",
      "title": "Send later, or on a schedule",
      "status": "available",
      "text": "The message box has a schedule mode: a message or a routine goes out at the time you set and wakes whoever it is for."
    },
    {
      "id": "widgets",
      "group": "Chats and work",
      "title": "To-dos that become tools",
      "status": "available",
      "text": "A to-do can turn into the widget it needs: a recorder, a teleprompter, a timer, a form. Answers can contain tables, charts and diagrams."
    },
    {
      "id": "side-tabs",
      "group": "Chats and work",
      "title": "Tabs beside the chat",
      "status": "available",
      "text": "A tab strip like Chrome’s beside every chat, for websites and for the app itself: My day, Work, Activity, Notes, Media, Files, Screen and an agent’s browser. Back and forward, history, zoom per site, mute, Ctrl+T and friends."
    },
    {
      "id": "wake",
      "group": "Proactive by design",
      "title": "Wake-ups and routines",
      "status": "available",
      "text": "A message, a task, a timer, a schedule or an event wakes an agent. It decides what to do, does it, and can plan the next step itself."
    },
    {
      "id": "tray",
      "group": "Proactive by design",
      "title": "Keeps working in the background",
      "status": "available",
      "text": "Close the window and the company keeps running from the system tray, as long as your computer is on."
    },
    {
      "id": "keep-running",
      "group": "Proactive by design",
      "title": "Keep agents running",
      "status": "available",
      "text": "Optional and off by default: Start with Windows or Start at login in Settings. The app starts when you sign in and comes back after a crash. On Windows it can also start for the next due work after you quit."
    },
    {
      "id": "verify",
      "group": "Proactive by design",
      "title": "Results checked against the goal",
      "status": "available",
      "text": "A task can carry a done-condition. Its result is then checked: does the named file exist, does the link answer, and a decision model reads the result without the agent’s conversation. Without a done-condition or a decision model the result stays marked unchecked."
    },
    {
      "id": "resume",
      "group": "Proactive by design",
      "title": "Picks work up again",
      "status": "available",
      "text": "After a lost connection, a used-up plan or a restart, a retry comes back with growing gaps and tells the agent what happened and what changed meanwhile. Tested with kills and restarts on a test engine, not yet with real providers."
    },
    {
      "id": "plans",
      "group": "The AI you already pay for",
      "title": "ChatGPT and Claude plans",
      "status": "available",
      "text": "Agents work through Codex with your ChatGPT plan and through Claude Code with a paid Claude plan. Add as many accounts per provider as you like. The app can install either tool for you and test it with a real answer."
    },
    {
      "id": "acp-plans",
      "group": "The AI you already pay for",
      "title": "Gemini CLI, Copilot CLI, Kimi Code",
      "status": "available",
      "text": "Built on the Agent Client Protocol: they sign in and run under the same rule as the other plans. Early: we have never seen them in a live run."
    },
    {
      "id": "routing",
      "group": "The AI you already pay for",
      "title": "The right model for each turn",
      "status": "available",
      "text": "Each turn gets an engine, an account, a model and an effort level. Live quota bars show how much of each plan is left, and you choose Best, Balanced or Cheaper."
    },
    {
      "id": "use-it",
      "group": "The AI you already pay for",
      "title": "Use it before you lose it",
      "status": "available",
      "text": "Weekly limits expire. Among accounts of the same priority, the plan quota that resets soonest while some is left goes first."
    },
    {
      "id": "switch",
      "group": "The AI you already pay for",
      "title": "Switch engines without losing context",
      "status": "available",
      "text": "When an agent moves from Codex to Claude Code or back, its session is rebuilt from a shared context ledger. Nothing has to be explained twice."
    },
    {
      "id": "lean",
      "group": "The AI you already pay for",
      "title": "Lean context",
      "status": "available",
      "text": "Tools load when they are needed instead of filling every call, and long old output can be compacted, hidden or brought back."
    },
    {
      "id": "apis",
      "group": "The AI you already pay for",
      "title": "API keys and local models",
      "status": "available",
      "text": "Optional: OpenAI, Anthropic, Google, OpenRouter, DeepInfra, Groq, Mistral, xAI, DeepSeek, Kimi, Qwen, or local servers like Ollama, LM Studio, llama.cpp and vLLM. Early: built, no live run recorded yet."
    },
    {
      "id": "uses",
      "group": "The AI you already pay for",
      "title": "A model for each use",
      "status": "available",
      "text": "One row per use (work, decisions, voice, speech, images, video, music, sound effects, embeddings), each with its provider and its model. A name the app does not know can simply be typed in."
    },
    {
      "id": "short-start",
      "group": "An open system",
      "title": "A short start",
      "status": "available",
      "text": "Connect one AI, type your name and company, press Start. The manager gets to know you and your work, and first asks how far it may go on this PC."
    },
    {
      "id": "browser-spaces",
      "group": "An open system",
      "title": "A browser per chat and per agent",
      "status": "available",
      "text": "Three places: the global browser where you work, a browser for each chat with tabs everyone in that chat shares, group chats too, and each agent’s own, which you can watch. Agents working in the background do not disturb your tabs."
    },
    {
      "id": "providers-file",
      "group": "An open system",
      "title": "Providers through a file",
      "status": "available",
      "text": "Every API connection is one entry in settings.json: a name, an address, a protocol, a key as an ID. A provider the app has never heard of works if it speaks one of the protocols. Say the provider’s name and hand over a key; the AI can write the rest."
    },
    {
      "id": "skills-files",
      "group": "An open system",
      "title": "Skills as open files",
      "status": "available",
      "text": "A skill is a folder with a SKILL.md, the format other agent tools use too. Each has a route: an agent’s own, its managers’, shared or built-in. The AI can read skills from folders of other tools when it chooses to, and writes new ones itself."
    },
    {
      "id": "memory-files",
      "group": "An open system",
      "title": "Memory as open files",
      "status": "available",
      "text": "One Markdown file per memory, in folders for the company, a branch or a single agent. Open them, change them, version them. Any change is searchable at once."
    },
    {
      "id": "own-connections",
      "group": "An open system",
      "title": "The AI builds its own connections",
      "status": "available",
      "text": "There is no WhatsApp button. Your agent sets it up: a small script next to the app, using the running app’s own interface. Telegram works the same way with a bot it creates. Other channels, such as Signal or Matrix, are the same kind of script: skills ship for WhatsApp and Telegram, and one that shows how for any other channel."
    },
    {
      "id": "settings-file",
      "group": "An open system",
      "title": "Settings in one file",
      "status": "available",
      "text": "settings.json holds every setting with a comment above it. Change a value and save; it applies within a second. So you can just tell your AI what to change."
    },
    {
      "id": "settings-pages",
      "group": "An open system",
      "title": "Settings pages on that file",
      "status": "available",
      "text": "General, Models, Voice, Calendar, Secrets, Apps, Skills & memory, Devices, Feedback and About, with a search. Each page opens the file behind it."
    },
    {
      "id": "file-tab",
      "group": "An open system",
      "title": "Files in a tab",
      "status": "available",
      "text": "A plain editor and a folder list beside the chat. Any path on your PC, Ctrl+S to save. Open settings.json, a skill, a memory or an agent’s instructions without leaving the app."
    },
    {
      "id": "see-prompt",
      "group": "An open system",
      "title": "See what is sent",
      "status": "available",
      "text": "Each agent’s folder keeps last-prompt.md: the whole system prompt the app sent on its last turn. Every shipped prompt can be replaced by your own version."
    },
    {
      "id": "protocol",
      "group": "An open system",
      "title": "Scripts and other apps",
      "status": "available",
      "text": "Everything the window can do, a program can do too: the wf command line, or plain HTTP to the running app with the token in runtime.json."
    },
    {
      "id": "theme",
      "group": "An open system",
      "title": "Change how it looks",
      "status": "available",
      "text": "theme.css on top of the app’s own styles, applied at once. Delete the file to go back. Your agent can do it for you."
    },
    {
      "id": "browser",
      "group": "Reaches your real tools",
      "title": "An agent’s own browser",
      "status": "available",
      "text": "Each agent can work in its own off-screen Chrome or Edge profile, with a live view you can watch and take over."
    },
    {
      "id": "computer",
      "group": "Reaches your real tools",
      "title": "Computer use",
      "status": "available",
      "text": "On Windows agents prefer real UI automation over raw clicks, and your own mouse and keyboard win. On Linux it runs over AT-SPI, with input on X11 only. On macOS agents get screenshots but cannot control the computer."
    },
    {
      "id": "android",
      "group": "Reaches your real tools",
      "title": "Your Android phone",
      "status": "available",
      "text": "Over ADB, by wireless debugging or USB: read the screen, tap, swipe and type on your own phone."
    },
    {
      "id": "apps",
      "group": "Reaches your real tools",
      "title": "Connected apps",
      "status": "available",
      "text": "MCP servers in one file, connectors/mcp.json, in the same format Claude Code and Cursor use. Agents find a server and write the entry themselves; the tools appear within a second."
    },
    {
      "id": "apps-page",
      "group": "Reaches your real tools",
      "title": "A page for your apps",
      "status": "available",
      "text": "Settings › Apps: a catalog of 17 services (Notion, Linear, GitHub, Zapier and more), add by address, add a mailbox, test, switch off, remove. It writes the same files."
    },
    {
      "id": "rest",
      "group": "Reaches your real tools",
      "title": "Any REST API",
      "status": "available",
      "text": "No connector needed. An agent calls the API directly with the key as an ID in the header, or writes a small script."
    },
    {
      "id": "email",
      "group": "Reaches your real tools",
      "title": "Email",
      "status": "available",
      "text": "Your own mailbox over IMAP and SMTP with an app password (Gmail, GMX, iCloud and others). A new mail can wake an agent. Outlook and Microsoft 365 are not supported yet."
    },
    {
      "id": "telegram",
      "group": "Reaches your real tools",
      "title": "Telegram",
      "status": "available",
      "text": "Mirror a chat to Telegram through your own bot and answer your agents from anywhere."
    },
    {
      "id": "vectors",
      "group": "Memory that stays",
      "title": "Search by meaning",
      "status": "available",
      "text": "Memory and any folder an agent chooses become searchable by meaning. A small local model does this on your PC, or an embedding provider you pick. Office documents can be read too."
    },
    {
      "id": "subconscious",
      "group": "Memory that stays",
      "title": "A subconscious beside each agent",
      "status": "available",
      "text": "A second process with a decision model and a stronger model looks through memory around a turn, notes what people state and tidies long context. What it finds reaches the agent as if the agent had looked it up itself. How reliable this is depends on the decision model."
    },
    {
      "id": "forget",
      "group": "Memory that stays",
      "title": "Forget",
      "status": "available",
      "text": "An agent forgets on request, and a memory is a file you can delete yourself."
    },
    {
      "id": "ledger",
      "group": "Memory that stays",
      "title": "Context ledger",
      "status": "available",
      "text": "Every part of an agent’s working context has a stable ID. Compaction happens only when needed, and what was hidden or compacted can be restored."
    },
    {
      "id": "voice",
      "group": "Voice, phone and more PCs",
      "title": "Dictation and calls",
      "status": "available",
      "text": "Dictate instead of typing. Call an agent inside a chat, optionally with OpenAI Realtime and your own key. Early."
    },
    {
      "id": "local-voice",
      "group": "Voice, phone and more PCs",
      "title": "Local speech",
      "status": "available",
      "text": "Speech-to-text and reading aloud with models that run on your own computer, downloaded when you pick them."
    },
    {
      "id": "phone",
      "group": "Voice, phone and more PCs",
      "title": "Your phone",
      "status": "available",
      "text": "Off until you turn it on. Scan a QR code and your PC serves the app to your phone in its browser, over Wi-Fi or Tailscale. No app store, no cloud account."
    },
    {
      "id": "remote",
      "group": "Voice, phone and more PCs",
      "title": "Several PCs",
      "status": "available",
      "text": "Pair your computers over the local network, a Remote ID or an address, compare a code, and let agents work on the machine that has the files."
    },
    {
      "id": "full-access",
      "group": "Access, secrets and the brake",
      "title": "Full access",
      "status": "available",
      "text": "Agents run without a sandbox and without approval prompts, with your rights. There is no permission gate, no guardian model and no money limit in the app. Read what that means before you install."
    },
    {
      "id": "first-question",
      "group": "Access, secrets and the brake",
      "title": "The first question",
      "status": "available",
      "text": "The first thing the AI asks is how far it may go: full access for good, this one search just this once, or less. It is a question, not a technical barrier."
    },
    {
      "id": "money-skill",
      "group": "Access, secrets and the brake",
      "title": "Spending, as a skill",
      "status": "available",
      "text": "How agents weigh a purchase is written in a skill with reference points and examples: roughly under 20 euros, above that, higher prices. A guide you can edit, not a limit in code."
    },
    {
      "id": "vault",
      "group": "Access, secrets and the brake",
      "title": "Secrets by ID",
      "status": "available",
      "text": "Passwords and keys are encrypted with your operating system’s protection. Agents work with IDs like secret://github, values stay out of prompts, and known values are masked in tool output. Every secret works everywhere; none is tied to a site."
    },
    {
      "id": "import-pw",
      "group": "Access, secrets and the brake",
      "title": "Password import",
      "status": "available",
      "text": "Import the export file from your browser or password manager once. Your agents then describe and organise the entries by ID, with the values out of their prompts. The export file itself is plain text: delete it afterwards."
    },
    {
      "id": "stop",
      "group": "Access, secrets and the brake",
      "title": "Stop all agent work",
      "status": "available",
      "text": "From the tray, the chat’s working list, the command palette or Ctrl+Alt+Shift+X. Every agent on this PC stops, and stays stopped after a restart until you resume. It does not undo what is done."
    },
    {
      "id": "journal",
      "group": "Access, secrets and the brake",
      "title": "Run journal",
      "status": "available",
      "text": "Tool calls are written down. A call whose effect is unknown after a crash is reported to the agent to check, not repeated blindly."
    },
    {
      "id": "local",
      "group": "Yours, on your computer",
      "title": "Local-first",
      "status": "available",
      "text": "Chats, tasks, memory, agents and the secrets catalog live on your PC. Prompts go to the AI providers you connect."
    },
    {
      "id": "backup",
      "group": "Yours, on your computer",
      "title": "Backup and restore",
      "status": "available",
      "text": "A copy of the data folder, without provider logins. Secret values stay in it encrypted for this PC, and agents’ browser profiles with their logged-in sites are copied too. Restoring happens at the next start and keeps a copy of what it replaced."
    },
    {
      "id": "export",
      "group": "Yours, on your computer",
      "title": "Your chats as files",
      "status": "available",
      "text": "A readable copy of every chat sits in the data folder, and Export writes a chat with its messages and files into a folder."
    },
    {
      "id": "repair",
      "group": "Yours, on your computer",
      "title": "A repair screen",
      "status": "available",
      "text": "If a start or an update fails repeatedly, the app opens a repair screen instead of leaving you with nothing."
    },
    {
      "id": "updates",
      "group": "Yours, on your computer",
      "title": "Quiet updates",
      "status": "available",
      "text": "On Windows and Linux a small icon appears when an update exists; one click installs it and restarts. On macOS the app replaces itself where macOS allows it."
    },
    {
      "id": "mac-update",
      "group": "Yours, on your computer",
      "title": "Mac updates without a new download",
      "status": "available",
      "text": "The Mac app is built to replace itself at an update. A Mac with a version from before the rename gets this one once by hand. It has never run on a real Mac yet, and it falls back to the manual download if macOS refuses."
    }
  ],
  "faq": [
    {
      "home": true,
      "q": "What is Autark?",
      "a": "Autark is a desktop app that gives you your own AI company: a manager, specialist agents and a private secretary. You talk to them in chats and drop tasks into one inbox. They pick up the work, delegate it among themselves, finish it and write it down, and they wake up on their own when something needs doing."
    },
    {
      "home": true,
      "q": "Is it free?",
      "a": "Yes. The app is a free download with no account and no subscription from us. The AI work runs on the plans you already have (for example a ChatGPT or Claude plan), on your own API keys or on local models, so you pay your AI providers directly."
    },
    {
      "home": true,
      "q": "Which AI subscriptions does it use?",
      "a": "ChatGPT plans through OpenAI Codex, and paid Claude plans through Claude Code, both via the official command-line tools of each provider. Gemini CLI, GitHub Copilot CLI and Kimi Code are built in too, but we have never seen them in a live run. The Claude Free plan cannot run agents."
    },
    {
      "q": "Do I need API keys?",
      "a": "No, not if you have a ChatGPT plan or a paid Claude plan. API keys and local model servers (Ollama, LM Studio, llama.cpp, vLLM) are optional and early: built, with no recorded live run yet. An AI provider the app does not know works if it speaks one of the supported protocols: it is one entry in settings.json."
    },
    {
      "home": true,
      "q": "What can agents do on my PC?",
      "a": "Everything your user account can. Agents run without a sandbox and without approval prompts: they can read and change your files, run commands, use a browser and every secret in the catalog. The first thing the AI asks is how far it may go, and that is a question, not a technical barrier. If a PC holds something you cannot risk, give the app its own Windows user or a virtual machine."
    },
    {
      "home": true,
      "q": "Can agents spend money or delete things?",
      "a": "Yes, technically. There is no money limit and no approval gate in the app. How agents weigh a purchase is written in a skill with reference points such as about 20 euros, and the rules ask them to check target, amount and recipient once before something they cannot undo. That is guidance you can edit, not a lock. An agent can pay with anything it can reach, such as a card you saved in the catalog or a shop where its browser is logged in."
    },
    {
      "home": true,
      "q": "Does my data go to your servers?",
      "a": "No. There are no Autark servers and the app has no analytics. Chats, tasks, memory, agents and the secrets catalog live on your own computer. Everything an agent works with goes to the AI provider you connected, including private files if it opened them. The access page lists every other place something leaves your PC."
    },
    {
      "q": "How are my passwords and keys handled?",
      "a": "They are encrypted with your operating system’s protection. Agents refer to them by ID, such as secret://github, and the app fills the value in at the last moment, so values stay out of prompts. Known values are masked in output. A command that receives a value can still do anything with it, and a password lying in a plain file on your disk is open to agents like any other file."
    },
    {
      "q": "Does Autark read my ChatGPT or Claude password or tokens?",
      "a": "The app does not. You sign in through the official command-line tools of each provider, each added account in its own folder, and Autark does not read, copy or forward those tokens. An agent with full access could open those folders like any others."
    },
    {
      "q": "How do I stop everything at once?",
      "a": "Use \"Stop all agent work\" in the tray, in the list behind the chat header’s live line or in the command palette, or press Ctrl+Alt+Shift+X while agents work. Every agent on this PC stops, and it stays that way after a restart until you resume. It does not reach work running on another PC or a script an agent left running on its own, such as a WhatsApp bridge, and it does not undo what was done."
    },
    {
      "home": true,
      "q": "Do agents keep working when I close the window?",
      "a": "Yes. By default the app keeps running in the system tray, and agents wake up on schedules, messages and events as long as your computer is on."
    },
    {
      "q": "What does \"proactive\" mean here?",
      "a": "Agents do not wait for a prompt. A schedule, a timer, a new task, a message or an event wakes them up. They decide what to do, do it, and can plan their own next step, for example a morning briefing, a follow-up email or a check on a running job. How much they do on a vague request still varies by model and by run."
    },
    {
      "q": "What happens on the first start?",
      "a": "A short start: a short tour, connect one AI, type your name and company, press Start. The manager then gets to know you and your work, first asks how far it may go, and afterwards points you to the two models it still needs, one for decisions and one for search."
    },
    {
      "q": "Can the AI change the app itself?",
      "a": "Yes, because the app is a set of open files: settings.json, skills, memory, connectors, theme.css and its prompts. Your agent edits them or calls the running app. It can add an AI provider, restyle the window or connect WhatsApp with a small script, without a connector being built for it first."
    },
    {
      "q": "Which platforms are supported?",
      "a": "Windows, macOS and Linux get the same version. Windows is where it is developed and tested most. On macOS agents get screenshots but cannot control the computer. On Linux computer control needs X11 for input, and the AppImage needs FUSE 2 and a system keyring."
    },
    {
      "q": "Why does Windows warn me about the installer?",
      "a": "The installer is not code-signed, so Windows SmartScreen shows “Windows protected your PC”. Choose \"More info\" and then \"Run anyway\". With Smart App Control turned on, Windows can block it outright. You can check the file against SHA256SUMS.txt from the same release."
    },
    {
      "q": "Why does my Mac refuse to open it?",
      "a": "The Mac app is signed ad hoc and not notarized, because there is no Apple Developer ID yet. On macOS 15 and newer, try to open it, choose Done, then go to System Settings, Privacy & Security, and choose Open Anyway. On older versions, Control-click the app and choose Open. If macOS says the app is damaged, remove the quarantine mark in Terminal: xattr -dr com.apple.quarantine \"/Applications/Autark.app\"."
    },
    {
      "q": "How do updates work?",
      "a": "The app checks GitHub Releases every minute while the window is open and every 15 minutes from the tray. On Windows and Linux a small icon appears when an update exists; one click downloads it, verifies it against the SHA-512 in the update manifest, installs it and restarts. On macOS the app replaces itself where macOS allows it, and otherwise the icon leads to the download. A Mac with a version from before the rename gets this one once by hand."
    },
    {
      "q": "Can I use it from my phone?",
      "a": "Yes. Turn on phone access in Settings, scan a QR code, and your PC serves the app to your phone’s browser over Wi-Fi or Tailscale. There is no app store app and no cloud relay of ours. Reaching it from outside your network takes Tailscale or your own setup; your agent can help with that."
    },
    {
      "q": "Can colleagues join my company?",
      "a": "Yes. Invite people by link or QR code. They work in the chats and tasks they belong to, next to your agents, and all data stays on your PC. This is early: it is built for one person first."
    },
    {
      "q": "What hardware do I need?",
      "a": "Autark is built and tested against an ordinary 8 GB RAM laptop. The heavy AI work happens at your AI provider, unless you choose a local model."
    },
    {
      "q": "Is Autark open source?",
      "a": "Not yet. The binaries are public on GitHub and an Apache-2.0 license is prepared, but the source code has not been published."
    },
    {
      "q": "Who makes Autark?",
      "a": "Teydra Lab from Salzburg, Austria, founded by Stanislaus Hees. Autark is part of Meta Jarvis, the Teydra Lab umbrella for projects around AI agents and the human who stays in charge of them."
    },
    {
      "q": "Was \"Proactive Agent Workforce\" the same product?",
      "a": "Yes. Proactive Agent Workforce was the working name. Updating from an older version moves the data folder to the new name, so there is no second app."
    }
  ],
  "updated": "2026-11-05"
}