Agency Orchestrator
Self-hosted AI agent orchestrator with visual workflows, Ollama support and cloud APIs. Runs on macOS, Windows, Linux or Docker under Apache 2.0.
Agency Orchestrator turns a task description into a team of specialist AI roles that divide the work and combine their results. It's for solo builders, developers and content creators who want repeatable workflows for product planning, research or code review. The open-source software uses the Apache 2.0 license and runs locally through a browser interface, a desktop app for macOS, Windows and Linux, or Docker on a server or NAS.
Its main distinction is the combination of ready-made roles and checks on deliverables. You can let AI assemble a team, choose roles yourself, or edit a workflow on a visual canvas without writing code. Independent tasks run in parallel, and later steps use earlier results. Workflows can include human approval gates, AI review against acceptance criteria and deterministic checks for missing files or required content. Failed checks trigger targeted rework.
Ollama provides local model inference. Cloud options include DeepSeek, Claude, OpenAI and OpenAI-compatible APIs; those routes send model requests to the selected provider. It can also reuse authenticated coding tools such as Claude Code and Codex without a separate API key. Credentials and generated files stay on your machine, but choosing a cloud backend doesn't make inference local. The project recommends larger Ollama models and cautions that small models can lose accuracy across role handoffs.
Runs can resume or accept revision feedback. MCP support lets Claude Code and Cursor invoke workflows, while self-contained HTML reports include each step's output and the team's work timeline.
