https://github.com/usezombie/usezombie
Autonomous Agent 🧟 Sleep is for humans.
https://github.com/usezombie/usezombie
agentic-ai agents autonomous-agents nullclaw openclaw zig
Last synced: 2 months ago
JSON representation
Autonomous Agent 🧟 Sleep is for humans.
- Host: GitHub
- URL: https://github.com/usezombie/usezombie
- Owner: usezombie
- License: mit
- Created: 2026-03-04T10:19:29.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2026-04-06T11:50:13.000Z (2 months ago)
- Last Synced: 2026-04-06T12:24:36.020Z (2 months ago)
- Topics: agentic-ai, agents, autonomous-agents, nullclaw, openclaw, zig
- Language: Zig
- Homepage: https://usezombie.com
- Size: 101 MB
- Stars: 2
- Watchers: 0
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: docs/contributing/SCHEMA_CONVENTIONS.md
- License: LICENSE
- Security: SECURITY.md
- Agents: AGENTS.md
Awesome Lists containing this project
README
**Heroku for agents, but the agent never sees your keys.**
**Run your agents 24/7. Walled, watched, killable.**
[](https://github.com/usezombie/usezombie/actions/workflows/test.yml?query=branch%3Amain)
[](https://codecov.io/gh/usezombie/usezombie/flags/apps)
[](https://github.com/usezombie/usezombie/releases)
[](https://usezombie.com)
[](https://docs.usezombie.com)
[](https://opensource.org/licenses/MIT)
---
> **Early Access Preview** · Pre-release — revised release coming up by April 11. APIs, CLI, and behavior may change without notice before general availability.
## Why we pivoted
We started by obsessing over one thing: making AI-generated code *correct*. Self-repair loops, quality scoring, scorecard evidence. Good problems — but narrow ones. Frontier models get better every quarter, and the gap we were optimizing for keeps shrinking on its own.
The tempting move is to double down — we built it, so we stick with it. But code is cattle, not pets. You don't keep a solution alive just because you wrote it.
```
____________________________
< code(human) === cattle now >
----------------------------
\ ^__^
\ (oo)\_______
(__)\ )\/\
||----w |
|| ||
```
So we killed the old approach and pivoted. UseZombie is now a runtime for always-on agents — you bring your agent, we handle the credentials (hidden from the sandbox, injected at the firewall), webhooks (wired automatically), audit logs (every action timestamped), and a kill switch. Your agent runs 24/7 without ever seeing a password.
## What zombies do now
- **Always-on agents** — your agent runs continuously in a sandboxed process, restarts on crash
- **Credentials hidden** — agents never see tokens; the firewall injects them per-request, outside the sandbox boundary
- **Webhooks wired** — receive events from email, Slack, GitHub, etc. without ngrok or custom servers
- **Observability** — see what your agent did, when, why, and how much it cost — before the invoice surprises you
- **Spend ceiling** — per-run token budgets and wall time limits; one bad prompt never becomes an infinite burn
- **Kill switch** — stop any agent mid-action from the CLI or web UI
## Stack
- **Auth** — [Clerk](https://clerk.com)
- **Hosting** — [Fly.io](https://fly.io) + bare-metal workers
- **Database** — [PlanetScale](https://planetscale.com) (Postgres)
- **Queue** — [Upstash](https://upstash.com) Redis Streams
- **Analytics** — [PostHog](https://posthog.com)
- **Email** — [Resend](https://resend.com)
## Documentation
| Resource | Description |
|----------|-------------|
| [docs.usezombie.com](https://docs.usezombie.com) | User-facing docs (guides, API reference, operator guide, contributing) |
| [playbooks/](playbooks/) | Agent-readable deployment playbooks and gate scripts |
## Playbooks
Playbooks are written to be understandable and executable by both humans and agents — each directory contains a markdown doc (`001_playbook.md`) and numbered shell scripts (`00_gate.sh`, `01_*.sh`, `02_*.sh`) that run in order.
See [playbooks/README.md](playbooks/README.md) for the full index and execution order.
## Local development
```bash
git clone https://github.com/usezombie/usezombie.git
cd usezombie
cp .env.example .env
make up # Start Postgres + Redis + zombied
make test # Run all unit tests
make lint # Format + lint
make doctor # Check config, Postgres, LLM key
```
See [docs.usezombie.com/contributing/development](https://docs.usezombie.com/contributing/development) for full setup.
## Agent support
This repo is configured for Claude Code, Codex, OpenCode, and Amp. All agents read the canonical [AGENTS.md](AGENTS.md) operating model.
| Agent | Config |
|-------|--------|
| Claude Code | `CLAUDE.md` → `AGENTS.md` |
| Codex | `.codex/instructions.md` |
| OpenCode | `.opencode/instructions.md` → `AGENTS.md` |
| Amp | `.amp/instructions.md` → `AGENTS.md` |
## License
MIT — Copyright (c) 2026 UseZombie