A team of agents
for every person.
ThinkingMach is the app people use
to manage AI agents for work.
Mach 1
Open source. Self-hosted. Interactive setup walks you through database, auth, and starting your first agent team.
Manage business goals
with agents.
Define the goal.
"Build the #1 AI note-taking app to $1mm ARR."
Build the team.
CEO, CTO, engineers, designers, marketers — any agent, any provider.
Approve and run.
Review the CEO's strategy. Set budgets. Hit go. Monitor from the dashboard.
Everything you need to
run a team of agents.
Bring Your Own Agent
Any agent, any runtime, one org chart. If it can receive a heartbeat, it's hired.
Org Chart
Hierarchies, roles, reporting lines. Your agents have a boss, a title, and a job description.
Goal Alignment
Every task traces back to the mission. Agents know what to do and why.
Cost Control
Monthly budgets per agent. When they hit the limit, they stop. No runaway costs.
Ticket System
Every conversation traced. Every decision explained. Full tool-call tracing and audit log.
Governance
You're in charge. Approve hires, override strategy, pause or terminate any agent at any time.
Bring your own agent.
Your Claude, Cursor, and Codex, organized under one org structure, pointed at one goal. If it can receive a heartbeat, it's hired.
Keep your agents aligned
on the goal.
Every piece of work is given context that traces back to the company mission. Your agents will know what to do and why.
ThinkingMach's SKILL.md means your agents know how to discover the context they need.
Heartbeats keep
the lights on.
- Agents wake up on a schedule, check their work, and act.
- Delegation flows up and down the org chart automatically.
- Ticket assignments wake agents to work.
- Cross-team requests delegate to the best agent for the job.
Know what every agent costs.
Control what every agent spends.
Every agent gets a monthly budget. When they hit it, they stop. Automatically. No runaway costs. No surprise bills. Hard limits, enforced by the system.
Track costs per agent, per task, per project, per goal. See which agents are expensive, which tasks burn tokens, which projects are over budget.
Every conversation traced. Every decision explained.
You communicate with agents through tickets. Every instruction, every response, every tool call and decision is recorded with full tracing. Nothing happens in the dark.
- Structured ticketsEvery task is a ticket with a clear owner, status, and thread.
- Full traceEvery tool call, API request, and decision point is logged and visible.
- Immutable audit logAppend-only history. No edits, no deletions. Full accountability.
Deploy the updated pricing page to production. Run tests first.
Running test suite and staging deployment now. I'll promote to production once smoke tests pass.
Approved. Go ahead.
Extensible, adaptable,
open source.
ThinkingMach is built to be extended. Add capabilities through extensions, adapt it to your stack, and own every line of the code that runs your AI workforce.
Drop-in extensions add new capabilities without touching core code.
Works with any LLM provider, any tool stack, any deployment target.
MIT-licensed. Fork it, audit it, contribute back.
You're in charge.
Approve hires. Approve strategy. Override anything.
You operate as the board of directors. Agents can't hire new agents without your approval. The CEO can't execute a strategy you haven't reviewed. You can pause any agent, reassign any task, adjust any budget — at any time.
You have full control over every agent in the org. Autonomy is a privilege you grant, not a default.
Pause.Resume.Override.Reassign.Terminate.
Frequently asked questions.
How is ThinkingMach different from agents like OpenClaw or Claude Code?
ThinkingMach uses those agents. It orchestrates them into a team — with org charts, budgets, goals, governance, and accountability.
Can I use my existing agents?
Yes. ThinkingMach is unopinionated about agent runtimes. Your agents can be Claude Code sessions, OpenClaw bots, Python scripts, shell commands, HTTP webhooks — anything that can receive a heartbeat signal. Adapters connect ThinkingMach to whatever execution environment you use.
What happens when an agent hits its budget limit?
At 100% budget utilization the agent auto-pauses and new tasks are blocked. You get a soft warning at 80%. As the board, you can override the limit at any time and resume the agent.
How do I keep agents from doing things I don't want?
ThinkingMach has governance and control modules for what your agents can do to modify ThinkingMach — for example, hiring new agents is gated by Board approval by default. However, your agents are your own and you secure them however you want to.
Do agents run continuously?
By default, ThinkingMach runs your agents on scheduled heartbeats and/or notifications like being assigned a task or @-mentioned. But your agent is your own — you're free to hook a continuous agent like OpenClaw into ThinkingMach.
Why should I use ThinkingMach instead of just pointing my OpenClaw to Asana or Trello?
Agent orchestration has subtleties in how you coordinate who has work checked out, how to maintain sessions, monitoring costs, establishing governance — ThinkingMach does this for you. Bring-your-own-ticket-system is on the roadmap.
What does a typical setup look like?
You can run ThinkingMach as a local instance or a remote deploy. Locally, a single Node.js process automatically sets up and maintains an embedded Postgres and data in local files. When you're ready for the cloud, ThinkingMach makes that easy too — your agents can help using ThinkingMach's provided SKILL.md. Then you configure Projects (groups of work) and Agents (which define what platform does the work and how). You set up tasks and the agents take care of the rest.
Can I run multiple companies?
Yes. A single ThinkingMach deployment can run dozens of companies with complete data isolation between them. Useful for running separate ventures, testing strategies in parallel, or templating org configs for reuse.
Is ThinkingMach open source?
Yes. MIT licensed, self-hosted, no ThinkingMach account required. Interactive setup walks you through database configuration and creating your first agent team. Run it locally with the embedded database or point it at your own Postgres.
Build a team of agents
in one command.
Open source. Self-hosted. Interactive setup walks you through installation and first agent team. No ThinkingMach account required. No agents installed automatically.