Questions about Agent Factory.
Updated Read as Markdown ↗
What Factory does, what it costs to run, and where the owner stays involved.
What is Agent Factory?
Agent Factory is a product-aware orchestration system for coding agents. It maps business capabilities, connects failures to root causes, coordinates bounded repairs and checks the released result. Aetha is building it as the next chapter of Orchestro.
Read the product overview →Does Factory replace my coding agent?
Factory coordinates your existing coding agents. Codex, Claude Code and Grok have repair-worker adapters. Muse Spark supports mapping proposals and Product Advisor reviews through the Muse CLI. The project keeps its issue history, evidence and permissions across supported agents.
Check supported roles →What can Factory monitor in an app?
Factory maps product capabilities such as onboarding, an AI assistant, server reliability and analytics. Each capability connects a user goal to screens, services, agents and data. Its success contract defines a metric, population, observation window, baseline, target and guardrails.
Explore app use cases →How does Factory recognize a bug that comes back?
It preserves the confirmed cause, attempted fixes, releases and independent checks. When a later failure reproduces the same cause, the issue can reopen with its earlier repair history attached. A similar complaint alone is not enough to establish recurrence.
Follow the repair example →Can the agent loop keep working for several days?
A configured watch process can keep checking and launching authorized work across days. It needs an awake host, a project schedule, authenticated CLIs and available quota. Individual worker runs still have limits; missing permission or evidence can require owner attention.
Set up scheduled operation →Does Factory require a separate model API bill?
The supported CLI workflow uses your coding-agent subscriptions rather than adding a separate model API billing layer. Model calls still consume subscription quota, and provider terms and limits apply. This is not zero-token or unlimited model execution. Early-access terms are discussed directly.
Understand providers and quotas →Does accepting an Advisor recommendation deploy a change?
No. An owner decision records intent and rationale. It does not grant code, merge or release permission. An explicit promotion can pass an accepted canonical-problem recommendation into governed intake after checking its scope and current evidence. That promotion still does not start execution.
See current availability →How do you distinguish a successful fix from a product improvement?
An independent check establishes whether the released repair works. Separate capability measurements compare product outcomes using the same definition and population. For example, onboarding activation might move from 62% to a proposed 64%. Before-and-after movement alone does not prove the repair caused it.
See metrics and guardrails →How does a coding agent add work to Factory?
The intake-work skill turns an owner request into a classified batch with acceptance criteria. The intake command validates the batch, checks for duplicates and returns stable backlog and decision IDs. Registering work does not start an agent. The CLI reference includes a complete JSON example.
Copy an intake example →Is Agent Factory publicly available?
Agent Factory is in private early access. Request repository access before following the quickstart. There is no public Factory npm installer or hosted Factory MCP endpoint advertised. The availability page separates implemented foundations from upcoming integrations and the full product-outcome pilot.
Read access requirements →Is Agent Factory the same as the old Orchestro MCP package?
Agent Factory replaces the Orchestro website and is the next step in the product. The legacy Orchestro MCP package remains a separate implementation. The website replacement does not migrate existing tasks, data, settings or MCP installations automatically.
Understand the transition →Are the dashboard examples real customer results?
No. The seven-day incident and capability measurements use illustrative data. Targets such as a two-percentage-point onboarding lift or 95% successful tool calls explain how product goals are defined; they are not customer benchmarks or proof of a completed production pilot.
Review implementation status →