Thrum
A local message bus that lets your coding agents review each other's work
Emerged finds new products worth a look, daily — what’s new today
If your agents drift on long jobs, a fixed, evidence-checked task plan that mixes models per step is a cheap way to keep smaller models on track.
Ordewell takes a single goal, researches your codebase and produces an ordered plan of tasks, each with its own runner, model, thinking effort and mode. It drives Claude Code, Codex or OpenCode using the subscriptions you already pay for, and marks tasks done from evidence rather than the agent's own say-so. It installs from npm or as a VS Code extension, with a tmux terminal UI. Apache-2.0, early, with no published benchmarks yet.
More like this
ZCode
Zhipu's open-source workbench for running AI agents on a codebase
Maxxwell
Desktop workspace for running and supervising a dozen coding agents at once
Halv
Run Claude Code, Codex, Kimi and GLM in one app that trims their token spend
Offrun
A Mac control room for Claude Code, Codex and other agents running side by side
Xencode
A Rust terminal coding agent that defaults to models on your own machine