Moadim
Moadim is an open-source, local-first loop engine for scheduling AI agents against repositories. It combines a prompt, schedule, and agent into repeatable routines that run unattended in isolated workbenches, with UI, REST, and MCP interfaces.
At a Glance
- Individual developers and open-source maintainers
- Software engineering teams using AI coding agents
- Local-first and privacy-conscious developers
- Teams managing recurring repository maintenance and CI-adjacent work
- +1 more
AI Tools by Moadim
(1)
Moadim
Scheduled AI Agent Loop Engine
Discussions
No discussions yet
Be the first to start a discussion about Moadim
Latest News
Moadim v3.2.8 added a built-in NanoClaw agent bridge and opt-in overlap sidecars for concurrent routine fires.
Moadim v3.2.7 added operator-confirmed manual routine runs while system power saving is active.
The project documented its positioning versus cron/systemd, GitHub Actions, Claude scheduling, and other agent runners, emphasizing local portable routine scheduling.
Moadim published the v3.0.0 daemon release during its rapid public release sequence.
Products & Services
A self-hosted Rust daemon that schedules agent routines—each a prompt, schedule, and agent—against repository workspaces. It runs locally on macOS and Linux and is distributed through Cargo, cargo-binstall, npm, Homebrew source builds, and GitHub release binaries.
A built-in dashboard for managing routines, viewing schedules and run history, inspecting logs, configuring settings and concurrency, and manually triggering runs.
Documented HTTP endpoints served by the daemon, including routine listing and management, with an OpenAPI schema and embedded Swagger UI.
An /mcp endpoint exposing routines as MCP tools so compatible agents can inspect, schedule, and re-fire their own routines.
Market Position
Moadim positions itself as a purpose-built scheduler for recurring AI-agent routines rather than a general script scheduler or hosted workflow platform. Its stated differentiators are local/offline execution, agent agnosticism, per-run isolation, git-trackable configuration, one-port UI/REST/MCP access, native MCP self-scheduling, and no cloud or per-minute billing. Its documented alternatives include cron/systemd timers, GitHub Actions, Claude Code scheduling, n8n, Temporal, Airflow, and standalone Codex/Hermes runners.
Leadership
Founders
Ofek Gabay
Creator and primary maintainer of Moadim (GitHub handle tupe12334). Public profile results identify him as a tech and AI enthusiast with experience as a Senior AI Engineer at Guidde, maintainer at Centy, Senior Software Engineer in the IDF Digital & Data unit, and software engineer at Amplication.
Executive Team
Ofek Gabay
Founder and primary maintainer
GitHub maintainer and author of the Show HN launch post; public profile results identify prior work at Guidde, Centy, the IDF Digital & Data unit, and Amplication.
Founding Story
Ofek Gabay introduced Moadim in a September 2026 Show HN post after building a scheduler to make agent routines git-compatible, agent-agnostic, open source, operating-system agnostic, multi-runner, and accessible through MCP, UI, and HTTP. The initial vision was a single source of truth in a repository for recurring agent work, rather than manually prompting agents or configuring low-level cron jobs.
Business Model
Revenue Model
Moadim is free, open source under the MIT license, self-hosted, and has no paid tier or account requirement. Users run it on their own machines and supply/configure their own agent tooling.
Pricing Tiers
MIT-licensed software; no paid tier and no sign-up. Installation is available via Cargo, cargo-binstall, npm, Homebrew source build, or release binaries.
Target Markets
- Individual developers and open-source maintainers
- Software engineering teams using AI coding agents
- Local-first and privacy-conscious developers
- Teams managing recurring repository maintenance and CI-adjacent work
- Operators who need agent scheduling without hosted orchestration infrastructure
- Recurring codebase maintenance and refactoring
- Scheduled bug detection, regression checks, and repository quality sweeps
- Automated pull/materialization and other periodic repository operations
- Daily messages, monitoring, and notifications generated by local agents
- Agent-driven workflows that need isolated workspaces and unattended execution
- Local/offline automation where a cloud queue or hosted runner is undesirable