Moadim.io: Open-source scheduler for agents supporting multiple runners and proto

What it does

Moadim is an open-source daemon that runs AI agents on a schedule. A user defines a loop—a prompt, a schedule, and an agent selection—and Moadim executes it at specified intervals. Each execution runs in an isolated workbench with a watchdog that kills hung processes. The tool supports multiple agents including Claude, Codex, Hermes, Pi, and NanoClaw. Loops can be triggered via REST API endpoints, MCP protocol, or a web UI.

Who it is for

Developers who need recurring agent tasks without cloud dependencies or manual intervention. This applies to teams running AI agents for monitoring, data processing, or repetitive automation where execution must survive reboots and remain entirely self-hosted. The tool targets those who want to treat agent scheduling like infrastructure code—manageable through Git repositories rather than UI configuration.

Pricing

Moadim is free and open source under the MIT license. There is no paid tier and no account creation required.

How it stands out

Moadim emphasizes local-first execution with no cloud queueing, hidden infrastructure, or vendor lock-in. Installation via Cargo or npm is straightforward, and the daemon runs on macOS and Linux using tmux for process isolation. The tool surfaces loops as REST endpoints with OpenAPI documentation and Swagger UI, making integration with other systems straightforward. Git-based configuration—defining routines as code that can be versioned and reviewed—differs from typical UI-driven schedulers. Support for multiple agents and protocol types (REST, MCP, HTTP) avoids single-vendor dependency.

What a founder should check

A builder considering this space should investigate whether existing cron alternatives, systemd timers, or container orchestration platforms (Kubernetes, Docker Compose) already solve this for their target users. Switching costs matter: teams already invested in cloud-native agent platforms or those requiring multi-machine scheduling may find the local-only constraint limiting. The moat here is thin—tmux-based isolation and REST/MCP bindings are straightforward to replicate. Verify whether price pressure exists: if competitors add cloud scheduling with similar agent support, Moadim's value proposition depends entirely on the local-first preference, which is not universal.

Thinking of building something like this?

Every launch here is a competitor to somebody's idea. If yours is close, check it against the market before you build: the Full Check names the rivals, the prices and the gaps.