HiveTerm
One workspace for all your AI agents
Start new thread
Ebrahim P. Leite

HiveTerm - One workspace for Claude, Codex, Gemini and your stack

by
HiveTerm
•
Running 5 AI agents across 3 projects? Tabs everywhere, no idea what each one is doing. HiveTerm is one terminal workspace where Claude Code, Codex, Gemini and your dev stack run side by side. Config-driven (hive.yml), per-agent recap in the sidebar, voice input, and an MCP server so agents can spawn and talk to each other. Free to start — up to 3 projects on the free tier. macOS, Windows, Linux.

Add a comment

Replies

Best
Ebrahim P. Leite
Maker
📌
Hey Product Hunt 👋 I launched HiveTerm here ~14 days ago as a way to manage dev processes from one place. Since then it became something I didn't expect: the workspace where I run every AI coding agent I use. What's new in this re-launch: - Voice input: push-to-talk straight into any agent - Per-agent recap line in the sidebar: no more tab-hopping to see what each agent is doing - Multi-agent: Claude Code, Codex, Gemini, all in split panes - MCP server (Queen): agents spawn sub-agents, talk to each other, project-isolated Free tier covers up to 3 projects. Works on macOS, Windows, Linux. Built solo, every feedback shapes the next release. Would love to hear: what's broken in your AI dev workflow today?
Austin
Exactly what I needed as well curious about usage logic though. Feels like it could burn through tokens. If it hits usage limits for one LLm will it handoff to the next?
Ansh Deb

OMG this is exactly what I needed. I use claude + codex workflow, so this is amazing.

Ng Jun Sheng

The per-agent recap line in the sidebar is the thing I didn't know I needed. Context-switching between Claude Code and Codex tabs to remember what each one was doing is genuinely one of the more annoying parts of running multiple agents. Does the MCP server let agents share context across projects or just spawn within the same project?

Ryuta Waku

Congrats on the re-launch, Ebrahim. Japan-based founder here.

This caught my eye because I’m also running Claude / Codex / Gemini across projects, and the “what is each agent doing?” problem is very real.

One Japan-specific thought: for AI coding-agent tools here, the adoption blocker may not be “can it run multiple agents?” but whether it stays understandable in real Japanese dev environments: Japanese docs, mixed JP/EN repo names, Japanese terminal output, internal specs, and privacy-sensitive codebases.

The local angle I’d test first is not just “one workspace for agents,” but “a safer multi-agent workspace for messy real codebases where context, logs, and process state stay readable.”