HiveTerm

One workspace for all your AI agents

55 followers

One workspace for all your AI agents

55 followers

Visit website

AI Coding Agents

•

Terminals

HiveTerm is a desktop terminal that runs your AI agents (Claude Code, Codex, Gemini, Cline, etc.) alongside servers, builds, and scripts, all orchestrated by a hive.yml file. What sets it apart is Queen, an integrated MCP server that allows agents to create sub-agents, read output from other processes, and send notifications to one another. Add to that per-process resource monitoring, Git integration with a diff viewer, automatic stack detection, and context-separated workspaces.

This is the 2nd launch from HiveTerm. View more

HiveTerm 0.16 - Voice

Launching today

Run every AI coding agent in one terminal - now by voice

HiveTerm turns your terminal into a workspace for AI coding agents. Run Claude Code, Codex, Gemini, Cursor side-by-side with your dev processes — split panes, config-driven via hive.yml, MCP server built in. New in v0.16: native voice input. Press Cmd+Shift+M, dictate your prompt to any agent, review before sending. Cross-platform mic capture (cpal, not the browser) with Whisper or Groq as your transcription provider. macOS, Windows, Linux. Built in Rust + React. hiveterm.com

Free Options

Launch tags:Productivity•Developer Tools•Artificial Intelligence

Launch Team

ElevenAgents by ElevenLabs — Scale conversations without scaling your team

Scale conversations without scaling your team

Promoted

HiveTerm

Maker

📌

Hey makers — Ebrahim here, founder of HiveTerm. Quick origin: I was constantly juggling 3-4 terminals — one for `npm run dev`, one for Claude Code, one for log tailing, one for a background sub-agent. Tabs got lost, restarts were manual, and switching context between my dev stack and my AI agents was friction every single hour. HiveTerm puts both in one place. You define your processes and your agents in a `hive.yml`, run `hv swarm`, and everything boots in split panes — including a built-in MCP server (the "Queen") that lets agents spawn sub-agents. We shipped the multi-agent workspace earlier this year. Today's launch is about voice. Press Cmd+Shift+M from anywhere and dictate your prompt to whichever agent is focused — Claude, Codex, Gemini, all of them. Native mic capture in Rust (via cpal, not WebView/getUserMedia), Voice Activity Detection that auto-stops on silence, OpenAI Whisper or Groq as your transcription provider, and per-language prompt biasing with dev vocabulary so it stops hallucinating "thanks for watching" on short audio (real bug we hit on quiet recordings). One deliberate choice: voice does NOT auto-submit. You always review the transcribed text before sending. Voice is for capture speed, not blind submission. Built in Rust + React + Tauri, runs on macOS, Windows, and Linux. hiveterm.com Would love your feedback — especially edge cases on Linux mic permissions, that's the platform I'm least confident on. Happy to answer anything in the thread.

Report

5h ago

Previous HiveTerm Launches

HiveTermOne workspace for all your AI agents

Launched on April 16th, 2026