Why Monkeys?

Why Choose Monkeys?

Monkeys is a workflow-centric AI application platform. It helps teams turn models, tools, data, agents, and human review steps into reusable applications that can be edited visually, executed through APIs, and deployed across cloud or private environments.

The current Monkeys ecosystem is split into multiple repositories. monkeys is the ecosystem and infrastructure entrypoint, while runtime source code lives in service repositories such as monkeys-server, monkeys-studio, monkeys-agent-server, workers, and tool services.

Workflow as the Product Surface

Monkeys treats workflows as the main abstraction. A workflow can include model calls, tool invocations, condition branches, loops, human tasks, data lookups, and output construction. The same workflow can power a chat view, an internal application, a scheduled task, or an API integration.

OpenAI-Compatible Integration

Workflows can be exposed through OpenAI-compatible endpoints. This makes Monkeys usable from existing chat clients, API gateways, and systems that already know how to call /v1/chat/completions style APIs.

Pluggable Tools

Tools are HTTP services described by a Monkeys manifest.json and an OpenAPI schema. Tool developers can use TypeScript, Python, Go, or any runtime that can expose HTTP APIs. Existing API services can also be imported through OpenAPI or manual configuration.

Team and Tenant Isolation

Data is organized by app, tenant, team, and user context. The core backend and capability services use prefixed tables, team headers, or tenant routing depending on the deployment shape.

Modular Services

The platform can run as a compact local stack or as a larger service set. Core product services can be combined with Agent, Data, Billing, Webhook, MCP, training, ComfyUI, and third-party tool services only when those capabilities are needed.

Operations Friendly

The local infrastructure stack includes PostgreSQL, Redis, Elasticsearch, Conductor, MinIO, Prometheus, and Grafana. Production deployments usually mirror the same boundaries with managed infrastructure and Kubernetes.

Monkeys is still moving quickly before a stable 1.0 release, so use repository README files and this documentation together when operating a specific version.