humux adapts to different workflows. Here are four common patterns — find yours.
Your AI pair programmer, researcher, and release manager
You're a developer who wants an AI that works with your codebase, not in a separate chat window. humux's coding harness lets the agent read, write, and edit files in a confined workspace. It can grep, refactor, run commands, and commit changes — all through Telegram, the CLI channel, or the admin UI.
< instead of <= — off-by-one, tokens still valid for 1 second after expiry get rejected.expires_at <= now()a1b2c3d ready to push.Full data sovereignty. Zero cloud dependencies.
Your AI should not send your emails, contacts, or calendar data to a third party. humux runs everything locally in a single Docker container on your own hardware. API keys are sealed in an encrypted vault — never exposed to the LLM context. Memory stays in your SQLite database. No telemetry, no phoning home.
Email triage, calendar management, task automation
Your inbox is overflowing, your calendar is fragmented across Google and iCloud, and you have recurring tasks that eat your time. humux reads and manages email (per-agent mailboxes), checks and creates calendar events, searches contacts, and runs scheduled tasks — all through Telegram, the CLI channel, or the admin UI.
Multiple agents, one Telegram group, zero collisions
Your team needs different AI personalities for different domains — a support agent, a research agent, a devops agent. humux's per-agent bots let you drop multiple agents into one Telegram group. Each has its own context, tools, and identity. They only reply when addressed and ignore other bots — no infinite loops, no confused replies.
humux works for all of them. One container, your configuration.