Launching today
yolonot
Permission firewall for Claude Code, Codex & 22 CLI agents
2 followers
Permission firewall for Claude Code, Codex & 22 CLI agents
2 followers
Open-source (Apache-2.0) command guardrails for 24 AI coding agents. Clears 54.2% of safe commands locally in ~1.6us via a Go Bash AST parser, inspects local scripts before execution, and cuts missed risks from 17.0% down to 4.9% (balanced) or 1.6% (strict).
Free
Launch Team / Built With
Render Sandboxes: a computer for every agentGive every agent its own machine. Starts in under a second
Promoted
Hey Product Hunt!
I built yolonot (https://yolonot.dev) because running CLI coding agents like Claude Code, Codex, Cursor, or OpenCode forces an annoying choice:
Manual prompts: Approve every "git status" and "pytest" by hand (~93% blind click-through fatigue).
--yolo mode: Give an LLM raw shell access (100% of risks missed).
Built-in auto modes: In our benchmarks (measured on Claude Code Auto Mode), it still missed 17.0% of risky commands, blocked 0.4% of safe ones with no ask state, and locked rules to a single tool.
yolonot is an open-source (Apache-2.0) Go binary that plugs into the hook layer of 24 coding agents:
Cuts missed risks down to 4.9% in balanced mode, 1.6% in strict mode, and 0.0% in paranoid mode across 456 benchmark cases (with 0.0% safe commands blocked).
Clears 54.2% of real-world commands locally in ~1.6 microseconds via a built-in Go Bash AST parser with zero API calls.
Reads and redacts local scripts on disk before execution (catching 14/14 Trojan scripts vs 0/14 by filename alone).
Works with fast cloud models, your existing Claude/Codex CLI login, or 100% local Ollama models (like Kev-4B).
Try it in 10 seconds:
GitHub: https://github.com/sezaakgun/yolonot
Would love your feedback and questions!