{
  "markdown": "<p align=\"center\">\n  <img src=\"doc/assets/G-HQ.jpeg\" width=\"600\" alt=\"Galyarder HQ\">\n</p>\n\n<h1 align=\"center\">Galyarder HQ</h1>\n\n<p align=\"center\">The control plane for AI-native companies.</p>\n\n<p align=\"center\">\n  <a href=\"LICENSE\"><img src=\"https://img.shields.io/badge/license-MIT-blue.svg\" alt=\"MIT License\"></a>\n  <a href=\"https://github.com/galyarderlabs/galyarder-hq/stargazers\"><img src=\"https://img.shields.io/github/stars/galyarderlabs/galyarder-hq\" alt=\"Stars\"></a>\n  <a href=\"https://github.com/galyarderlabs/galyarder-framework\"><img src=\"https://img.shields.io/badge/framework-galyarder-black\" alt=\"Framework\"></a>\n</p>\n\n<p align=\"center\">\n  Open source · Self-hosted · No account required\n</p>\n\n---\n\nThe next generation of companies won't be built by bigger teams. They'll be built by founders who figured out how to make AI work like a company — not just a tool.\n\nGalyarder HQ is the infrastructure for that. A self-hosted control plane where you define the org, set the goals, and let agents run 24/7 through heartbeats, queues, budgets, approvals, and review gates. Not a chatbot. Not a workflow builder. A company — with structure, accountability, and governed execution.\n\nThe shift is already happening. The question is whether you're running it, or still babysitting it.\n\n---\n\n## The idea\n\nMost people use AI agents the same way they used to use Google — one question at a time. That's not leverage. That's just a faster keyboard.\n\nReal leverage is when your agents have context, continuity, and coordination. When they know the company goal, not just the task. When they report to someone, not just to you. When they work while you sleep.\n\nThat's what Galyarder HQ is built for.\n\n---\n\n## How it works\n\nYou build an org. Departments, roles, reporting lines. You hire agents — Claude, Codex, Cursor, Gemini, or anything that speaks HTTP. You assign goals and let work flow down the hierarchy.\n\nAgents wake on heartbeats. They check their queue, pick up tasks, execute, and report back. You review what matters. With the right hosting and budget, the company does not sleep; it just stays bounded by the rules you set.\n\nEvery decision is logged. Every cost is tracked. Every agent has a budget. When they hit the limit, they stop. You're always in control — you just don't have to be present.\n\n---\n\n## What you get\n\n**Structure, not chaos**  \nOrg charts. Departments. Reporting lines. Your agents have titles, managers, and a reason to exist beyond the last message you sent them.\n\n**Continuity across sessions**  \nTasks are tickets. Conversations are threaded. Sessions persist. Your agents remember what they were doing — even after a reboot.\n\n**Cost discipline**  \nMonthly budgets per agent. Hard stops when limits are hit. You'll never wake up to a surprise bill again.\n\n**Governance without micromanagement**  \nApproval gates for what matters. Config versioning. Rollback. You set the rules once. The system enforces them.\n\n**One control plane, many companies**  \nRun multiple AI-native companies from a single deployment. Complete data isolation. One command surface for your entire portfolio.\n\n---\n\n## Quickstart\n\n```bash\ngit clone https://github.com/galyarderlabs/galyarder-hq.git\ncd galyarder-hq\npnpm install\npnpm dev\n```\n\nOpen **http://localhost:3100**\n\nNo database setup needed — uses embedded PGlite in dev.\n\n**Requirements:** Node.js 20+, pnpm 9+\n\n---\n\n## Agent Framework\n\nGalyarder HQ pairs with [Galyarder Framework](https://github.com/galyarderlabs/galyarder-framework) — 40 specialized agents and 132 production-ready skills for AI assistants.\n\nInstall Framework in your AI tool. Connect it to HQ. Your agents now have a company to work in.\n\n---\n\n## Supported agents\n\nClaude Code · Codex · Cursor · Gemini · OpenCode · Any HTTP agent\n\nIf it can receive a heartbeat, it's hired.\n\n---\n\n## FAQ\n\n**What does a typical setup look like?**  \nA single Node.js process, an embedded database, and your agents. For production, point it at your own Postgres and deploy wherever you like. The agents handle the rest.\n\n**Can I run multiple companies?**  \nYes. Unlimited companies, complete data isolation, one deployment.\n\n**How is this different from just using Claude Code directly?**  \nClaude Code does the work. Galyarder HQ runs the company that Claude Code works for. Org charts, budgets, goals, governance — the infrastructure that makes a team of agents actually function like a team.\n\n**Do agents run continuously?**  \nOn heartbeats and event triggers by default. You bring the agent. HQ coordinates.\n\n---\n\n## Development\n\n```bash\npnpm dev           # Start dev server (API + UI)\npnpm build         # Build everything\npnpm typecheck     # Type check\npnpm test:run      # Run tests\npnpm db:generate   # Generate migrations\npnpm db:migrate    # Apply migrations\n```\n\nFull guide: [doc/DEVELOPING.md](doc/DEVELOPING.md)\n\n---\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md).\n\n---\n\n## License\n\nMIT © 2026 Galyarder Labs\n\n---\n\n<p align=\"center\">\n  The future of work isn't more people. It's better infrastructure.<br>\n  <br>\n  Open source. Self-hosted. Built for founders who think in systems.\n</p>\n",
  "bytes": 5133,
  "sha": "b108ce5f733fbba38d836dad17876c93b73eed73bd8f645a76bb6afed424e28b",
  "repo_slug": "galyarderlabs/galyarder-hq",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_galyarderlabs_galyarder_hq_01932e79/readme"
}