nono

Sandboxing tool that restricts the filesystem, network and command access of AI coding agents.

Used to sandbox AI coding agents so they can safely run in YOLO or permission-skipping mode,1234 with policy-based blocking, proxy credential injection and rollbacks.567 The most common caveat is that writing per-project policies can be cumbersome,89 and one daily user notes it is not a perfect sandbox.10
The HN verdict, written from 10 cited comments
5recommend
1warn against
May 8, 2026first mention
Oct 21, 2026latest mention
Nov 25 – Oct 26
Used forfor AI coding agentsfor teamsfor Claude Code usersfor macOS userssandboxing agentssandboxing coding agents

Why people recommend it

Safe YOLO mode · 3

Gives commenters enough confidence to run agents without one-by-one approval prompts.

“Nono has been my daily driver since the start of the year.”

“I use nono.sh for sandboxing”

Show 1 more

“The external sandboxing tool I use (nono) supports rollbacks for this kind of situation.”

Action blocking · 2

Blocks agent actions that can't be prompted around, enforced outside the agent harness.

“We use nono [1] for this internally, essentially blocking any actions we don't want agents to be able to do without human approval.”

“The flexibility of a good nono setup (which is easy because they ship a nono skill that claude can consume) actually locks things down significantly better”

Easy policies · 2

Tries to make policy writing easy and ships a nono skill that Claude can consume.

“Nono does try to make policy writing easy”

“The flexibility of a good nono setup (which is easy because they ship a nono skill that claude can consume) actually locks things down significantly better”

Credential injection · 1

Its proxy injects real API credentials so the agent only sees dummy ones.

“For my LLM agents, I use a tool called Nono, whose macOS implementation is built on the same foundation as sandbox-exec.”

Rollbacks · 1

Supports rolling back destructive changes an agent makes.

“The external sandboxing tool I use (nono) supports rollbacks for this kind of situation.”

No microVM · 1

Avoids a microVM, so already-installed tooling works inside the sandbox.

“Nono does try to make policy writing easy”

Caveats and dissent

Policy upkeep · 3

Every project seems to need new access rules, and one user found limiting commands cumbersome.

“Nono does try to make policy writing easy”

“I tried nono to limit commands an agent can run but that proved to be cumbersome.”

Show 1 more

Imperfect isolation · 1

Default network rules allow escape via a global tmux server.

“Nono has been my daily driver since the start of the year.”

Who recommends it in their own words

From the makers not counted in the ranking

“I wanted to introduce a project I have building for the past few weeks”

All comments 12

“For my LLM agents, I use a tool called Nono, whose macOS implementation is built on the same foundation as sandbox-exec.”

“I tried nono to limit commands an agent can run but that proved to be cumbersome.”

“Nono has been my daily driver since the start of the year.”

“The flexibility of a good nono setup (which is easy because they ship a nono skill that claude can consume) actually locks things down significantly better”

“I'll be sandboxing agents (with tools like nono, smolmachines, etc) going forward”

“We use nono [1] for this internally, essentially blocking any actions we don't want agents to be able to do without human approval.”

“I use nono.sh for sandboxing”

“but also have seen nono: https://github.com/always-further/nono”

“Nono does try to make policy writing easy”

“The external sandboxing tool I use (nono) supports rollbacks for this kind of situation.”

Threads where it came up

Also written as Nono, nono.sh. Data: nono.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.

Keyboard shortcuts
j / kNext / previous item
o or EnterOpen selected item
/Search
g hHome
g tTop
g rRising
g cCategories
tToggle dark mode
?This help