#agent-loop

Gives agents a think→act loop — plus recursive decomposition (RLM): break a hard task down, fan out, verify, synthesize. Goal in, coordinated actions out. Replaces LangChain, CrewAI, AutoGen. Zero required deps.

Terminal-based, Agent-friendly Ticketing System

Fresh-context iterative loop for Claude Code — Leader/Worker/Verifier protocol inspired by Ralph Loop and Codex design-desk

A small, nestable agent-loop job primitive: loop(body→until→review) composed with dag, pluggable engines, agent-judge gates, budget + run records. TypeScript.