Feature Catalog

Every capability,
documented and deliberate

humux ships with a complete set of integrated capabilities. No plugins to buy. No cloud subscriptions. Everything runs locally in one container.

Per-agent Telegram bots

Each agent runs its own Telegram bot. Add several to one group — they take turns, never loop, only reply when addressed. Each has isolated context.

Voice messages

Send voice notes in Telegram. humux transcribes with faster-whisper and responds by voice via edge-tts or Kokoro.

Inline approvals

Approve or reject write actions directly from Telegram inline buttons. Permission rules use glob patterns: ALWAYS, ASK, NEVER.

Reactions

Acknowledge messages with emoji reactions instead of text — thumbs-up, heart, eyes. Pre-approved, no prompt interruption.

Per-chat settings

Gate who can trigger an agent in each group and who may DM it: everyone, nobody, or specific user IDs.

WhatsApp (wacli)

Read and send WhatsApp messages via the wacli CLI tool. Integrated as a skill the agent can learn.

CLI channel

Chat with the agent directly from your terminal — SSH in or run uv run python -m core.cli. Resumable sessions, no Telegram needed.

Full email access

Read, compose, search, and manage emails. Powered by Himalaya CLI, a battle-tested Rust MUA.

Per-agent mailboxes

Each agent can own a dedicated mailbox or be granted read / read-write access to your inbox. Bind accounts per agent.

Vault-secured credentials

Credentials resolve from the encrypted vault — they never enter the model's context. Bitwarden import supported.

Read & write events

Check your schedule, create events, update existing ones. CalDAV works with Google Calendar, iCloud, Fastmail, and any CalDAV server.

Per-agent access levels

Bind calendar accounts per agent with read or read-write permissions. Granular control over who sees and modifies what.

Scheduled briefings

Morning briefings include today's events, emails, and reminders. Cron-based scheduling for any custom task.

Contact search & creation

Search contacts, create new entries. Supports CardDAV (Purelymail, iCloud, Fastmail) and Google Contacts via the contacts CLI.

Per-agent scoping

Bind contact books per agent with read or read-write access. Agents only see what you grant them.

Multiple agents

Create and switch between agent identities. Each has its own character, skill/tool scope, voice, and account bindings. Enable or disable per agent.

Two-tier memory

Permanent long-term facts and expiring short-term context. Both extracted automatically from conversations via LLM inference.

Subagents

Delegate scoped subtasks to a sub-loop. Run sync (result in-turn) or background (distilled summary). Set step/token/thinking budgets.

Markdown skills

Teach the agent new capabilities by writing markdown files — no code required. Add tool integrations by describing the CLI command.

Scheduled jobs

Cron-based scheduling for morning briefings, email checks, contact sync, and custom tasks. Filter by agent in the admin UI.

Context usage meter

See token count and percentage of context window used in the Inspect tab. Manage context budget per agent.

Setup wizard

Step-by-step first-boot configuration. Walks through API keys, Telegram bot setup, account connections, and agent creation.

Agent editor

Configure agent character, skills, tools, voice, Telegram bot, and account bindings. Tabbed interface with sys prompt, inference, and provider settings.

Log streams

Per-agent log streams filterable by stream, level, time, and text. Inspect context, token usage, and chain-of-thought.

Memory inspection

Browse and manage long-term and short-term memory entries. See what your agent remembers and edit or delete entries.

Job management

View, create, and cancel scheduled jobs. Monitor subagent runs. Filter jobs by agent.

Secrets vault

Encrypted two-tier vault: infrastructure keys (machine-key sealed) and agent secrets (admin-password sealed). Bitwarden import.

Speech-to-text

Powered by faster-whisper. Transcribe voice messages from Telegram. On-device, no external API needed.

Text-to-speech

edge-tts for cloud-quality voices, or Kokoro 82M for fully offline multilingual TTS. Both integrated and selectable.

Web search

Tavily integration for real-time web search. The agent can look up current information, news, and research topics.

Browser automation

Optional headless Playwright browser. Read JS-heavy pages, act on websites with persistent logged-in profiles. Per-domain approval.

Web artifacts

The agent publishes pages and sites under /artifacts/ using workspace file tools. Sandbox CSP for security. Shareable links.

Encrypted secrets vault

Two-tier encryption: infrastructure keys sealed with machine key, agent secrets sealed with admin password. Values never enter model context.

Granular permissions

Glob-pattern rules: ALWAYS allow, ASK for approval, NEVER allow. Interactive approval via Telegram inline buttons for write actions.

Safe coding harness

File tools confined to workspace directory. Read pre-approved, writes ask first. Path traversal blocked. Per-call command approval.

Multi-provider support

Choose from Anthropic Claude, OpenAI, Grok (xAI), Google Gemini, DeepSeek, or OpenRouter. Configure per-agent or globally. Image generation via OpenRouter, fal.ai, or OpenAI.

Provider sub-tabs

Admin UI splits LLM config into Inference, Providers, and History sub-tabs. URL-based navigation for copy-pasteable links.

Ready to run your own AI?

One container. Your data. Your rules.