Moadim
1 commentThe first launch from Moadim
Moadim
A scheduler that runs your coding agents on a loop
#1 of the day · 16 Sep 2026
Moadim is an open-source loop engine for AI agents: a loop is a prompt, a schedule and an agent — Claude Code, Codex, Hermes or Pi — and on every tick it runs that agent against your repository in a fresh, isolated workbench with a watchdog that kills hung runs. It ships as a single Rust binary with its own scheduler, so no cron daemon, and every loop is also a documented REST endpoint and an MCP tool, so agents can schedule other agents.
Meet the team
What people who use it say
Nobody has reviewed Moadim yet. If you have used it, you are the one who can say whether it holds up.
Write the first review
Sign in to comment
Once an agent can do a task on its own, the next question is who runs it every morning, and the answer so far has been a shell script and a cron line that nobody remembers writing.
A scheduler that knows what an agent run is — isolated workspace, a timeout, a session to reap — is the missing piece, and exposing each loop as an MCP tool means the agents can manage the schedule themselves, which is either the point or the problem depending on your temperament.