no_human
Open-source agent that takes a ticket to a reviewed pull request, with proof
Emerged finds new products worth a look, daily — what’s new today
It stops coding agents from declaring victory early by binding each task to runnable verification commands, so you spend less time re-checking their output.
Unlazy is an open-source skill for coding agents such as Claude Code and Codex that enforces completion discipline. Its Depth Tree method decomposes a task into layers and gives each leaf the full time budget, so effort compounds with depth. The core mechanism is gate contracts: checkpoints with executable verification commands, expected outcomes and captured evidence that must pass before work is accepted. It adds scoped pipelines for parallel work, approval records, a Claude Code Stop hook that blocks until gates pass, and a gate-strength linter. Zero-dependency Node.js.
More like this
DashClaw
Approval gates and an audit ledger for coding agents left running unattended
Thrum
A local message bus that lets your coding agents review each other's work
ZCode
Zhipu's open-source workbench for running AI agents on a codebase
captain-hook
Turns the corrections you keep giving Claude Code into typed, tested hooks
Maxxwell
Desktop workspace for running and supervising a dozen coding agents at once