One daemon, reading what your tools already wrote
Start the daemon
The tray app or bun run start, then localhost:7787. No console window.
It finds what is already there
Claude transcripts, Codex rollouts, the OpenCode database, every isolated instance. Nothing to import.
Reply, queue, or hand it over
Reply to a Claude session, queue claude runs, or let your agents drive it over MCP.
Three tools, and an honest line between them
All three histories, one list. Only Claude takes instructions back, because only Claude ships a CLI you can hand a prompt.
| What you can do | Claude | Codex | OpenCode |
|---|---|---|---|
| Read and live-tail the conversation | Yes | Yes | Yes |
| Search and filter alongside everything else | Yes | Yes | Yes |
| Type back into the session | Yes | No | No |
| Message several sessions at once | Yes | No | No |
| Queue and schedule runs | Yes | No | No |
| Isolated instances, per account | Desktop + CLI | Desktop + CLI | Not applicable |
| Remaining quota per account | Yes | No | No |
| Open, download or copy the raw file | Yes | Yes | Rendered from its database |
Scroll the table sideways for Codex and OpenCode.
The part your terminal window is missing
One list, three tools
Every conversation, newest first, badged by tool. Filter, click, watch it stream.
Isolated Claude instances
Every Claude Desktop profile with its account, plan and live usage. Open, focus, quit, rename, recolour.
Codex gets its own windows
Its own CODEX_HOME per instance, so work and personal OpenAI logins run side by side.
A composer that reaches the session
Type at the bottom of a transcript and it reaches that session. Send, queue, schedule, or fan out to several.
A queue you can walk away from
Each run carries its own prompt, directory, model and account. Queue it, rate-limit it, or set it for 3am.
Runs that survive a restart
A detached supervisor, so quitting the app does not kill what is in flight.
A rate limit is not a dead end
Sessions stopped by a 5-hour limit resume when the window resets. Gated on weekly usage, and off by default.
Handing a task to ChatGPT
Packs the task and repo into a bounded Markdown file and opens ChatGPT. Secrets left out; you press submit.
Quick instance mode
--instances skips everything but a start/focus/stop window, on its own port.
Built for you and your agents
You get a dashboard
Sessions, instances, queue and accounts, in your regular browser.
Your agents get MCP
The whole API on MCP stdio, so Claude Code, Claude Desktop or Cursor drive the exact state you click.
get_session
tail_session
run_queue_item
get_run_events
set_scheduler
set_monitor
launch_instance
quit_instance
create_codex_instance
launch_codex_instance
check_my_usage
usage_budget
{
"mcpServers": {
"agenthydra": {
"command": "bun",
"args": ["run", "--cwd", "<path-to-agenthydra>", "mcp"]
}
}
}Agents that know when to stop
An agent can read an account's remaining quota before a heavy fan-out, and route work to whichever account has the most left.
Pace the fan-out, don't crash into the wall
Install
v0.43.0 · all releases
Or from source with Bun. Either way the dashboard is at localhost:7787, loopback only.
AGENTHYDRA_FAKE=1 swaps in a harmless stand-in, so nothing touches your quota or repos.
The claude CLI for dispatch; Claude Desktop for instances. Codex and OpenCode optional.
Tray launcher is Windows only; macOS and Linux builds are unverified. Instance management needs the classic Claude Desktop installer, not MSIX.
Local by default, with one exception: a periodic update check sends the app version, a coarse OS tag and a random install id. Never an IP, hostname, path, account or email. AGENTHYDRA_NO_PING=1 opts out. MIT.
What people weigh AgentHydra against
Separate terminal windows
One terminal each, alt-tabbed, history scattered. Putting it in one list is the whole point.
Crystal (now Nimbalyst)
Spins up parallel sessions in fresh git worktrees. AgentHydra spins up nothing; it reads what you already have.
Conductor
Mac-only worktree launcher for parallel agents. AgentHydra runs everywhere and reads existing sessions.
ccmanager
A terminal UI for worktree sessions. AgentHydra is a browser dashboard plus MCP, with a queue and quota checks.
Stop alt-tabbing to remember what is running
One daemon. Every session in one list, every instance in one table, an MCP door for your agents.
Download for WindowsQuestions people actually ask
Is AgentHydra free?
Yes. AgentHydra is free and open source under the MIT license, and no account is needed to run it. There's no cloud service behind it and nothing to sign up for: it reads the local Claude, Codex and OpenCode stores already on your machine and serves a dashboard on localhost.
What AI coding tools does AgentHydra support?
AgentHydra reads and lists sessions from Claude Code, Codex and OpenCode side by side, newest first, each badged with the tool that wrote it. Claude Code sessions come from your transcripts, Codex from its rollout JSONL files, and OpenCode CLI and Desktop share one local SQLite store, so both show up automatically.
Can I reply to a Codex or OpenCode session from AgentHydra?
No, only Claude sessions can be messaged back. AgentHydra can read and live-tail Codex and OpenCode conversations just like Claude's, but only Claude ships a CLI that can be handed a prompt, so Codex and OpenCode sessions open without a reply box instead of a send button that quietly does nothing.
Is AgentHydra private, or does it send my data anywhere?
Not entirely. The one thing that leaves your machine on its own is a periodic update check against studio.connections.icu: it sends the app version, a coarse OS tag, and a random per-install id, and the server derives a coarse location (country, region, city, timezone), your network's ASN, locale, and a truncated user agent from that request, but never an IP address, hostname, username, file path, account, or email. Set AGENTHYDRA_NO_PING=1 to opt out; the check then goes straight to GitHub's API, with no install id or telemetry. Otherwise AgentHydra is one daemon on loopback with your regular browser and a tray icon: no cloud service, no account to sign up for. The other thing that leaves is the optional ChatGPT handoff, which opens the provider and leaves the final upload to you.
What can my AI agents do through AgentHydra's MCP server?
AgentHydra exposes its whole API over MCP stdio, so Claude Code, Claude Desktop or Cursor can drive the same state as the dashboard: sessions, the run queue, the scheduler, the auto-resume monitor, accounts, quota, and the Claude, CLI and Codex instance managers. Agents can also check an account's remaining quota before fanning out work.
Does AgentHydra work offline?
Yes, for reading and messaging: AgentHydra needs no network connection to list, tail, search or reply to sessions, since it only reads local transcript, rollout and SQLite files already on your machine. The claude CLI calls it dispatches still need whatever connectivity Claude itself requires, and the optional update check and ChatGPT handoff are both online-only and skippable.
How is AgentHydra different from tools like Crystal, Conductor or ccmanager?
Those tools launch new Claude Code or Codex sessions in fresh git worktrees so you can run several tasks in parallel. AgentHydra instead reads the sessions you already have across Claude Code, Codex and OpenCode in one browser dashboard, and adds a queue, scheduler, instance manager and MCP server on top, without needing a new worktree per session.
Do I need Bun installed to use AgentHydra?
No. Bun is only needed to run AgentHydra from source (bun install, bun run build, bun run start). The packaged Windows .exe, and the one-executable Linux and macOS builds on the Releases page, bundle everything and need nothing else installed.