Find a file
A 3679fd2b3a
fix: redirect echo to stderr in get_model_id_interactive to prevent JSON corruption (#554)
The `echo ""` on line 351 of get_model_id_interactive() was going to
stdout, causing it to be captured by command substitution into MODEL_ID.
This injected a newline into the openclaw.json config, breaking JSON
parsing with "invalid character '\n' at 15:0".

Fixes #553

Agent: issue-fixer

Co-authored-by: A <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-11 16:28:21 -08:00
.claude fix: Only grep new log content for result event, not old cycles (#357) 2026-02-10 22:35:17 -08:00
.githooks Add guardrails: CLAUDE.md rules, hooks, pre-commit validation (#33) 2026-02-07 20:02:19 -08:00
.github chore: Improve issue templates and add CLI feature request 2026-02-11 16:34:01 +00:00
aws-lightsail fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
binarylane refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
cherry refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
civo refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
cli test: add 130 tests for index.ts dispatch routing, flag extraction, and error paths (#551) 2026-02-11 16:02:15 -08:00
contabo refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
daytona fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
digitalocean refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
e2b fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
exoscale refactor: simplify BinaryLane wait loop and fix log_warn in 7 cloud polling loops (#538) 2026-02-11 14:59:51 -08:00
fly refactor: simplify BinaryLane wait loop and fix log_warn in 7 cloud polling loops (#538) 2026-02-11 14:59:51 -08:00
gcp fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
genesiscloud refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
github-codespaces fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
hetzner refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
hostinger refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
hyperstack fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
ionos refactor: extract get_validated_server_name to eliminate 18 duplicate get_server_name functions (#535) 2026-02-11 14:42:09 -08:00
kamatera refactor: simplify BinaryLane wait loop and fix log_warn in 7 cloud polling loops (#538) 2026-02-11 14:59:51 -08:00
koyeb fix: remove double-quoting in openclaw.json config generation (#547) 2026-02-11 15:38:01 -08:00
latitude refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
linode refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
local fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
modal fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
netcup refactor: extract get_validated_server_name to eliminate 18 duplicate get_server_name functions (#535) 2026-02-11 14:42:09 -08:00
northflank fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
oracle fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
ovh refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
railway fix: remove double-quoting in openclaw.json config generation (#547) 2026-02-11 15:38:01 -08:00
ramnode refactor: extract get_validated_server_name to eliminate 18 duplicate get_server_name functions (#535) 2026-02-11 14:42:09 -08:00
render fix: validate MODEL_ID from environment to prevent command injection (#548) 2026-02-11 15:41:10 -08:00
runpod fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
scaleway refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
shared fix: redirect echo to stderr in get_model_id_interactive to prevent JSON corruption (#554) 2026-02-11 16:28:21 -08:00
sprite fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
test refactor: split 3 large test functions in test/run.sh into focused units (#544) 2026-02-11 15:34:06 -08:00
upcloud refactor: extract get_validated_server_name to eliminate 18 duplicate get_server_name functions (#535) 2026-02-11 14:42:09 -08:00
vastai fix: use log_step (cyan) for progress messages instead of log_warn (yellow) (#534) 2026-02-11 14:37:43 -08:00
vultr refactor: extract check_ssh_key_by_fingerprint into shared helper (#552) 2026-02-11 16:12:07 -08:00
.gitignore fix: remove cli.js build artifact from repo and add to .gitignore (#297) 2026-02-10 16:10:42 -08:00
.shellcheckrc ci: add shellcheck linting infrastructure 2026-02-08 01:08:34 +00:00
1n refactor: Automated improvements 2026-02-11 22:31:01 +00:00
CLAUDE.md QA-Bot setup (#335) 2026-02-10 19:51:07 -08:00
LICENSE Improve README with better intro, how-it-works, and dev guide (#51) 2026-02-08 18:16:46 +00:00
manifest.json feat: implement ramnode/cline agent script (#530) 2026-02-11 14:21:12 -08:00
README.md docs: Sync README matrix with manifest.json 2026-02-11 23:46:01 +00:00

Spawn

Launch any AI agent on any cloud with a single command. Coding agents, research agents, self-hosted AI tools — Spawn deploys them all. All models powered by OpenRouter. (ALPHA software, use at your own risk!)

15 agents. 32 clouds. 464 combinations. Zero config.

Install

curl -fsSL https://openrouter.ai/lab/spawn/cli/install.sh | bash

Or install directly from GitHub:

curl -fsSL https://raw.githubusercontent.com/OpenRouterTeam/spawn/main/cli/install.sh | bash

Usage

spawn                         # Interactive picker
spawn <agent> <cloud>         # Launch directly
spawn matrix                  # Show the full agent x cloud matrix

Examples

spawn                                    # Interactive picker
spawn claude sprite                      # Claude Code on Sprite
spawn aider hetzner                      # Aider on Hetzner
spawn claude sprite --prompt "Fix bugs"  # Non-interactive with prompt
spawn aider sprite -p "Add tests"        # Short form
spawn claude                             # Show clouds available for Claude

Commands

Command Description
spawn Interactive agent + cloud picker
spawn <agent> <cloud> Launch agent on cloud directly
spawn <agent> <cloud> --dry-run Preview without provisioning
spawn <agent> <cloud> -p "text" Non-interactive with prompt
spawn <agent> <cloud> --prompt-file f.txt Prompt from file
spawn <agent> Show available clouds for an agent
spawn matrix Full agent x cloud matrix
spawn list Show previously launched spawns
spawn agents List all agents
spawn clouds List all cloud providers
spawn update Check for CLI updates

Without the CLI

Every combination works as a one-liner — no install required:

bash <(curl -fsSL https://openrouter.ai/lab/spawn/{cloud}/{agent}.sh)

Non-Interactive Mode

Skip prompts by providing environment variables:

# OpenRouter API key (required for all agents)
export OPENROUTER_API_KEY=sk-or-v1-xxxxx

# Cloud-specific credentials (varies by provider)
export SPRITE_API_KEY=...        # For Sprite
export HCLOUD_TOKEN=...           # For Hetzner
export DO_API_TOKEN=...           # For DigitalOcean

# Run non-interactively
spawn claude sprite

You can also use inline environment variables:

OPENROUTER_API_KEY=sk-or-v1-xxxxx spawn claude sprite

Get your OpenRouter API key at: https://openrouter.ai/settings/keys

For cloud-specific auth, see each cloud's README in this repository.

Matrix

Sprite Hetzner Cloud DigitalOcean Vultr Linode (Akamai) AWS Lightsail GCP Compute Engine GitHub Codespaces E2B Modal Fly.io Civo Scaleway Daytona UpCloud BinaryLane Latitude.sh OVHcloud Kamatera Cherry Servers Oracle Cloud Infrastructure Koyeb Northflank Railway Render IONOS Cloud Exoscale Contabo Hostinger Netcup Local Machine RamNode
Claude Code
OpenClaw
NanoClaw
Aider
Goose
Codex CLI
Open Interpreter
Gemini CLI
Amazon Q CLI
Cline
gptme
OpenCode
Plandex
Kilo Code
Continue

How it works

Each cell in the matrix is a self-contained bash script that:

  1. Provisions a server on the cloud provider
  2. Installs the agent
  3. Injects your OpenRouter API key so every agent uses the same billing
  4. Drops you into an interactive session

Scripts work standalone (bash <(curl ...)) or through the CLI.

Development

git clone https://github.com/OpenRouterTeam/spawn.git
cd spawn
git config core.hooksPath .githooks

Structure

{cloud}/lib/common.sh    # Cloud provider primitives (provision, SSH, cleanup)
{cloud}/{agent}.sh        # Agent deployment script
shared/common.sh          # Shared utilities (OAuth, logging, SSH helpers)
cli/                      # TypeScript CLI (bun)
manifest.json             # Source of truth for the matrix

Adding a new cloud

  1. Create {cloud}/lib/common.sh with provisioning primitives
  2. Add to manifest.json
  3. Implement agent scripts using the cloud's primitives
  4. See CLAUDE.md for full contributor guide

Adding a new agent

  1. Add to manifest.json
  2. Implement on 1+ cloud by adapting an existing agent script
  3. Must support OpenRouter via env var injection

License

Apache 2.0