# Orchestro

> Orchestro is the control plane for autonomous software agents. It maintains persistent AI-agent loops with durable state, work selection, recovery, independent verification, production feedback and learning. Your models supply intelligence and agent harnesses execute; Orchestro keeps the work and evidence connected. Super private beta, primarily used on our own internal projects. The preview is free with hands-on onboarding; users supply their own agents, subscriptions and infrastructure. Public self-service access and broader unattended autonomy remain in development.

Private early access. Live Factories uses public-safe observations with freshness. Story labels historical cases; product illustrations and operating examples use sample data, not customer benchmarks. CLI model calls consume subscription quota. Check availability before describing roadmap items as implemented.

## Pages
- [Home](https://www.orchestro.org/): Give your project a goal. Orchestro runs persistent AI-agent loops with local execution, independent verification and durable history. Explore the real Factory.
- [How the Factory works](https://www.orchestro.org/how-it-works): See how a shared kernel governs work, evidence and independent checks. Reliability and Content add their own tools, policies and definitions of done.
- [Live Factories](https://www.orchestro.org/live): See public observations from real Orchestro Factories. Check executor status, durable progress and blockers, with clear privacy and freshness boundaries.
- [Interactive recovery demo](https://www.orchestro.org/live/demo): Try a 64-second software reliability simulation. See false completion rejected, interrupted work resumed and the released outcome checked. No signup.
- [From Reliability to Content](https://www.orchestro.org/story): From ARVO reliability lessons to a real Content Factory: why Orchestro keeps work, evidence and history beyond the chat, with local execution and cloud visibility.
- [Developer preview](https://www.orchestro.org/early-access): Apply for a free private Orchestro pilot with one software project and one recurring problem. Hands-on setup; bring your own agents and infrastructure.
- [Request privacy](https://www.orchestro.org/privacy): How Orchestro handles your preview application, project description and contact details, and how to update or remove a request.
- [Illustrative operating examples](https://www.orchestro.org/docs/operating-examples): Explore the retained reliability and marketing scenarios: bounded experiments, competing evidence, verification and governed work. All scenarios are illustrative.
- [Introduction](https://www.orchestro.org/docs): Orchestro is the control plane for autonomous software agents. It maintains persistent AI-agent loops with durable state, work selection, recovery, independent verification, production feedback and learning. Your models supply intelligence and agent harnesses execute; Orchestro keeps the work and evidence connected. Super private beta, primarily used on our own internal projects. The preview is free with hands-on onboarding; users supply their own agents, subscriptions and infrastructure. Public self-service access and broader unattended autonomy remain in development.
- [Factory lifecycle](https://www.orchestro.org/docs/factory/lifecycle): A report in the backlog is not a promised PR. Follow classification, shared problems, selection, root cause, verification, evidence waits and the distinction between Backlog Zero and Operational Zero.
- [Quickstart](https://www.orchestro.org/docs/quickstart): Set up Orchestro, check your coding CLI and connect a project in observation mode.
- [How the loop works](https://www.orchestro.org/docs/how-it-works): How Orchestro connects product mapping, recommendations, owner decisions, agent work and verification.
- [Your loop](https://www.orchestro.org/docs/loops): One kernel maintains different missions. See the historical Reliability and Content pilots, the current Content observation, and the private Domain Builder foundation for public self-service setup.
- [Work, waits & evidence](https://www.orchestro.org/docs/work-lifecycle): Follow 12 signals into actionable work, evidence waits and independent verification. See incremental sample targets, return conditions and shared-cause history.
- [Coding agents](https://www.orchestro.org/docs/agents): Compare Codex, Claude Code and Grok repair workers with Muse Spark mapping, Advisor and Governor roles. Learn about CLI subscriptions, quotas and permissions.
- [Agent tools](https://www.orchestro.org/docs/tools): Real CLI examples for project status, backlog, Governor, cloud snapshots and Advisor, plus the documented command catalog for intake, mapping and verification.
- [MCP & integrations](https://www.orchestro.org/docs/mcp): Use the private hosted MCP control plane and owner cockpit. Understand project-scoped reads, cloud intent, local execution and remaining chat-client acceptance.
- [App use cases](https://www.orchestro.org/docs/use-cases): Define practical goals for onboarding conversion, AI tool-call reliability, server errors and analytics performance, with baselines and guardrails.
- [Grow](https://www.orchestro.org/docs/grow): A real Content Factory pilot on the Reliability kernel. Evidence, editorial reviews, publishing and verification, with general campaign automation still in development.
- [FAQ](https://www.orchestro.org/docs/faq): Answers about Orchestro: supported coding agents, subscriptions, multi-day loops, recurring bugs, intake, early access, custom loops and growth work.
- [What’s available](https://www.orchestro.org/docs/roadmap): The private cockpit and hosted MCP are deployed. Content has a public read model. Separate historical proof, implemented capabilities and unfinished self-service or chat-client acceptance.
- [Name & history](https://www.orchestro.org/docs/orchestro): Orchestro is the product name. Learn how the legacy MCP package and the northstar command fit in.

## Plain-text references
- [Full site content](https://www.orchestro.org/llms-full.txt): generated from the visible page content.
- [Command catalog](https://www.orchestro.org/commands.json): 36 CLI commands in the September 15 catalog snapshot; not hosted MCP tools. Newer read commands are documented in Agent tools.
- [Home in Markdown](https://www.orchestro.org/index.md)
- [How the Factory works in Markdown](https://www.orchestro.org/how-it-works.md)
- [Live Factories in Markdown](https://www.orchestro.org/live.md)
- [Interactive recovery demo in Markdown](https://www.orchestro.org/live/demo.md)
- [From Reliability to Content in Markdown](https://www.orchestro.org/story.md)
- [Developer preview in Markdown](https://www.orchestro.org/early-access.md)
- [Request privacy in Markdown](https://www.orchestro.org/privacy.md)
- [Illustrative operating examples in Markdown](https://www.orchestro.org/docs/operating-examples.md)
- [Introduction in Markdown](https://www.orchestro.org/docs.md)
- [Factory lifecycle in Markdown](https://www.orchestro.org/docs/factory/lifecycle.md)
- [Quickstart in Markdown](https://www.orchestro.org/docs/quickstart.md)
- [How the loop works in Markdown](https://www.orchestro.org/docs/how-it-works.md)
- [Your loop in Markdown](https://www.orchestro.org/docs/loops.md)
- [Work, waits & evidence in Markdown](https://www.orchestro.org/docs/work-lifecycle.md)
- [Coding agents in Markdown](https://www.orchestro.org/docs/agents.md)
- [Agent tools in Markdown](https://www.orchestro.org/docs/tools.md)
- [MCP & integrations in Markdown](https://www.orchestro.org/docs/mcp.md)
- [App use cases in Markdown](https://www.orchestro.org/docs/use-cases.md)
- [Grow in Markdown](https://www.orchestro.org/docs/grow.md)
- [FAQ in Markdown](https://www.orchestro.org/docs/faq.md)
- [What’s available in Markdown](https://www.orchestro.org/docs/roadmap.md)
- [Name & history in Markdown](https://www.orchestro.org/docs/orchestro.md)