agentleFS
OpenRouterTeamOpenRouterTeam/spawn/CLAUDE.md

spawn

CLAUDE.md251 starsChanged 5 months agoApache-2.0View on GitHub

Spawn is a matrix of agents x clouds. Every script provisions a cloud server, installs an agent, injects OpenRouter credentials, and drops the user into an interactive session. manifest.json is the source of truth. It tracks: - agents — AI agents and self-hosted AI tools (Claude Code, OpenClaw, Codex CLI, ...) - clouds — cloud providers to run them on (Sprite, Hetzner, ...) - matrix — which cloud/agent combinations are "implemented" vs "missing" All cloud provisioning and agent setup logic…

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. OpenRouterLabs/spawn★ 251100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository9 other files it gives its agents.

The file

We split this CLAUDE.md into the rules it gives an agent

6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All cloud provisioning and agent setup logic lives in TypeScript under packages/cli/src/.

    line 56
  2. Very specific:

    Update manifest.json matrix status to "implemented"

    line 66
  3. Very specific:

    Update the cloud's sh/{cloud}/README.md with usage instructions

    line 67
  4. Distinctive:

    Commit with a descriptive message

    line 68
  5. Distinctive:

    When you encounter bugs, stale references, broken functionality, or architectural issues that are outside the scope of your current task, …

    line 72
  6. Distinctive:

    Do NOT silently ignore problems: If you find something weird and won't fix it now, file an issue so it's tracked.

    line 84

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Spawn

Spawn is a matrix of agents x clouds. Every script provisions a cloud server, installs an agent, injects OpenRouter credentials, and drops the user into an interactive session.

The Matrix

manifest.json is the source of truth. It tracks:

  • agents — AI agents and self-hosted AI tools (Claude Code, OpenClaw, Codex CLI, ...)
  • clouds — cloud providers to run them on (Sprite, Hetzner, ...)
  • matrix — which cloud/agent combinations are "implemented" vs "missing"

File Structure

spawn/
  packages/
    cli/
      src/index.ts               # CLI entry point (bun/TypeScript)
      src/manifest.ts            # Manifest fetch + cache logic
      src/commands/              # Per-command modules (interactive, list, run, etc.)
      src/commands/index.ts       # Barrel re-export of all command modules
      package.json               # npm package (@openrouter/spawn)
  sh/
    cli/
      install.sh                 # One-liner installer (bun → npm → auto-install bun)
    shared/
      github-auth.sh             # Standalone GitHub CLI auth helper
      key-request.sh             # API key provisioning helpers (used by QA)
    e2e/
      lib/*.sh                   # E2E helper libraries
    test/
      macos-compat.sh            # macOS compatibility test script
    {cloud}/
      {agent}.sh                 # Agent deployment scripts (thin bash → bun wrappers)
      README.md                  # Cloud-specific usage docs
  .claude/
    rules/                       # Modular rules (auto-loaded by Claude Code)
    scripts/                     # Hook scripts (enforce-worktree, validate-file, pre-merge-check)
    skills/setup-agent-team/
      trigger-server.ts          # HTTP trigger server (concurrent runs, dedup)
      discovery.sh               # Discovery cycle script
      refactor.sh                # Dual-mode cycle script (issue fix or full refactor)
      start-discovery.sh         # Launcher with secrets (gitignored)
      start-refactor.sh          # Launcher with secrets (gitignored)
  .github/workflows/
    discovery.yml                # Scheduled + issue-triggered discovery workflow
    refactor.yml                 # Scheduled + issue-triggered refactor workflow
  manifest.json                  # The matrix (source of truth)
  fixtures/                      # API response fixtures for testing
  README.md                      # User-facing docs
  CLAUDE.md                      # This file — project overview

Architecture

All cloud provisioning and agent setup logic lives in TypeScript under packages/cli/src/. Agent scripts (sh/{cloud}/{agent}.sh) are thin bash wrappers that bootstrap bun and invoke the CLI.

sh/shared/github-auth.sh — Standalone GitHub CLI installer + OAuth login helper. Used by packages/cli/src/shared/agent-setup.ts to set up gh on remote VMs.

sh/shared/key-request.sh — API key provisioning helpers sourced by the QA harness (qa.sh) for loading cloud credentials from ~/.config/spawn/{cloud}.json.

After Each Change

  1. bash -n {file} syntax check on all modified scripts
  2. cd packages/cli && bunx @biomejs/biome check src/ — must pass with zero errors (lint + format) on all modified TypeScript
  3. Update manifest.json matrix status to "implemented"
  4. Update the cloud's sh/{cloud}/README.md with usage instructions
  5. Commit with a descriptive message

Filing Issues for Discovered Problems

When you encounter bugs, stale references, broken functionality, or architectural issues that are outside the scope of your current task, file a GitHub issue immediately rather than ignoring them or trying to fix everything at once:

gh issue create --repo OpenRouterLabs/spawn --title "bug: <brief description>" --body "<details>"

Examples of when to file:

  • Dead code or stale references to files/functions that no longer exist
  • Broken features (e.g., spawn delete references non-existent shell scripts)
  • Security concerns that need separate review
  • Architectural debt that would be too large to fix in the current PR

Do NOT silently ignore problems. If you find something weird and won't fix it now, file an issue so it's tracked.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.