DashClaw
Approval gates and an audit ledger for coding agents left running unattended
Emerged finds new products worth a look, daily — what’s new today
If an agent in your repo ran a destructive command last Tuesday, this is the cheapest way to learn which one, when, and whether policy allowed it.
Shield KYA is an MIT-licensed CLI that sits between a coding agent and its tools, logs each call to a local dashboard, and refuses a short default list: destructive shell commands, secret reads, exfiltration. It hooks into Claude Code, Cursor, Codex, Gemini CLI and about twenty other hosts, and installs with npm or Homebrew. The open part records and blocks the basics. Approval queues, policy as code and signed evidence bundles live in a hosted workspace with a free tier and paid plans from $29 a month. The GitHub organisation behind it, The Pixel Boys, names nobody.
More like this
Geiger
Inventory every AI agent on your machine and what it can reach
CTRLRun
An open-source safety layer that checks every AI agent action before it runs
OpenBot
Self-hosted AI coworkers, each with its own computer and a policy gateway
ZCode
Zhipu's open-source workbench for running AI agents on a codebase
no_human
Open-source agent that takes a ticket to a reviewed pull request, with proof