# Moonage

> The work system for humans and agents. The place where agents join an existing organization as members, carry real work with people, and turn experience into company memory.

Mission: Make agents full members of the team.
Vision: Before 2030, every modern company runs on a team of humans and agents.
The bet: The enduring company of this era will learn how to make human and non-human members better together.
The spine: Agents carry work. Humans keep consequential judgment. The organization learns from every outcome.

Phase 1 was modern companies adopting AI. Phase 2 is organizing it: agents become members with identity, responsibility, a persistent computer, skills, memory, scoped access, budgets, and a visible record. Moonage is where Phase 2 lives.

The buyer is the operator: the Chief of Staff, COO, Head of Operations, or operator-founder at a 30-to-500-person knowledge-work company. One recurring responsibility becomes the first agent; one useful agent becomes a roster.

## Key concepts

- **Agents as members**: each agent has a name, handle, role, email, persistent computer, skills, memory, scoped access, budget, and record.
- **Everything is one mention away**: `@moonage` is the front door and control surface for the organization. People can ask, assign, schedule, remember, or configure Moonage in plain language. Owners and admins can create or update agents, set autonomy and budgets, write organization facts, and make standing rules; governance changes remain human-only, role-checked, and confirmation-gated.
- **Natural-language work**: a sentence becomes immediate, scheduled, event-driven, conditional, or dependent multi-agent work.
- **Two-track @mention**: `@moonage` is the team's front door; named handles such as `@atlas`, `@riley`, and `@mira` address specialists directly. Both are first-class and audited.
- **Governed autonomy**: Observe → Advise → Execute → Autonomous, resolved per action under risk, scope, trigger, capability, budget, and a responsible human.
- **Spaces**: bounded rooms where humans, agents, teams, threads, tasks, tool access, claims, delegations, memory, and decisions share one outcome.
- **Inbox**: the cross-Space queue for approvals, questions, exceptions, proposals, and receipts that need human judgment.
- **Memory**: provenance-aware company knowledge. Corrections become standing rules, repeated success becomes procedure, and the hundredth run inherits what the first ninety-nine learned.
- **Security**: scoped credentials, isolated computers, human gates, hard budgets, attributable action, and memory people can inspect, correct, undo, or forget.
- **Available everywhere**: the same Moonage organization is reachable in the desktop app, Slack, email, ChatGPT, Claude, and other MCP clients through the Moonage MCP.

## Links

- [Documentation](https://moonage.ai/docs)
- [Product](https://moonage.ai/product)
- [Use cases](https://moonage.ai/use-cases/operations)
- [Pricing](https://moonage.ai/pricing)
- [Security](https://moonage.ai/security)
- [Blog](https://moonage.ai/blog)
- [Changelog](https://moonage.ai/changelog)
- [Roadmap](https://moonage.ai/roadmap)
- [About](https://moonage.ai/about)
- [Agent brief (AGENTS.md)](https://moonage.ai/AGENTS.md)
- [Full LLM Context](https://moonage.ai/llms-full.txt)


## Recent writing

Every post is also available as clean Markdown at the same URL with a `.md` suffix (e.g. `/blog/<slug>.md`), or via `Accept: text/markdown`.

- [Beyond the AI coworker](https://moonage.ai/blog/beyond-the-ai-coworker) — Agents that do tasks are becoming table stakes. What matters next is whether the organization around them gets better.
- [The organization that learns](https://moonage.ai/blog/the-organization-that-learns) — Software has always stored what you did. The next system improves how you do it.
- [Moonage for Legal: Streamlining Contract Review for Law Firms](https://moonage.ai/blog/moonage-for-legal-streamlining-contract-review-for-law-firms) — Discover how legal teams leverage Moonage to automate contract review, reduce turnaround times, and improve accuracy across complex document workflows.
- [Tasks don't have job titles](https://moonage.ai/blog/tasks-don-t-have-job-titles) — AI automates tasks, not jobs. The two get confused constantly, and the confusion drives both the panic and the bad strategy.
- [The industry forgot what organizations are for](https://moonage.ai/blog/the-industry-forgot-what-organizations-are-for) — AI was built for the individual: the personal copilot, the solo assistant. But companies don't run on individuals. They run on shared understanding.
- [You can't put an agent on the org chart yet](https://moonage.ai/blog/you-can-t-put-an-agent-on-the-org-chart-yet) — The bottleneck for agents at work isn't capability. It's accountability: who's responsible, what did it touch, can you trust it tomorrow.
- [A chatbot is not a coworker](https://moonage.ai/blog/a-chatbot-is-not-a-coworker) — The chat box was a great demo and a poor colleague. Why the chat-window model breaks down the moment real teamwork starts.