nono is a sandbox that gives AI agents controlled execution paths for running commands with limited privileges. It is used to let coding agents work while protecting credentials and production systems.
Borrowing it
Nothing to install: this file belongs to nolabs-ai/nono. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/nolabs-ai/nono/main/AGENTS.mdgit clone --depth 1 https://github.com/nolabs-ai/nonoWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/nolabs-ai/nono/agents-md)<a href="https://agentmods.dev/instructions/nolabs-ai/nono/agents-md"><img src="https://agentmods.dev/badge/instructions/nolabs-ai/nono/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/nolabs-ai/nono/agents-md"><img src="https://agentmods.dev/badge/instructions/nolabs-ai/nono/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00945 | $0.00945 |
| Opus 5.5 | $0.00378 | $0.00378 |
| Sonnet 5.5 | $0.00189 | $0.00189 |
| Haiku 4.5 | $0.00094 | $0.00094 |
Grade A, and why
nono AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
nono agent guide
nono is a security-critical, capability-based sandbox for untrusted AI agents. It uses Landlock on Linux and Seatbelt on macOS. Treat every change as a security change until its affected boundary is understood.
Architecture
This Cargo workspace contains five packages:
crates/nono/: policy-free sandbox primitive andCapabilitySet.crates/nono-cli/: policy, profiles, execution, diagnostics, and UX.crates/nono-proxy/: network filtering and credential injection.bindings/c/: C FFI and generatednono.hheader.crates/nono-test-support/: unpublished CLI test support.
The library applies only capabilities a caller supplies. Policy—default
protected paths, profile semantics, command policy, prompts, and UX—belongs in
nono-cli, not the library. Preserve that boundary.
Non-negotiable security rules
- Fail secure: configuration, validation, and enforcement failures must deny access or return an error; never silently weaken enforcement.
- Grant least privilege. Keep filesystem read/write permissions distinct and avoid broad directory, network, or command grants.
- Use
Pathcomponent comparison, never string prefix comparison, for paths. Validate and canonicalize paths at the enforcement boundary; consider symlink and TOCTOU behavior. - Treat paths, environment variables, profiles, packages, registry responses, command arguments, URLs, headers, and agent-generated files as untrusted.
- Escape and validate data interpolated into Seatbelt profiles. Linux Landlock is allow-list based and cannot express deny-within-allow, so do not create a broad grant that defeats a required denial.
- Review Linux and macOS separately when a change affects enforcement, networking, paths, processes, sockets, credentials, or diagnostics.
- Keep credentials out of logs, errors,
Debug, audit output, and child environments. Usezeroizefor secret material in memory.
Implementation and verification
- Use
NonoErrorandResultfor expected failures. Do not add production.unwrap()or.expect(); restrictunsafeto justified low-level/FFI boundaries and document each block with// SAFETY:. - Use checked, saturating, or overflowing arithmetic where overflow affects a security decision.
- New capability or sandbox behavior needs tests. Tests that modify
HOME,TMPDIR,XDG_CONFIG_HOME, or related variables must restore them and keep the modified window short. - Regenerate
bindings/c/include/nono.hafter changingbindings/c. - Run focused checks while working. Before a PR or when handing off a material
code change, run
make ci; usemake fmtto format. - Before changing release tooling, workflows, or package configuration, read the release runbook.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · -130 lines · -1,561 tokens per session 816dcb2d44cb
- 1mo ago Changed · +7 lines · +97 tokens per session 1bc899507358
- 1mo ago First seen · 206 lines · 2,409 tokens per session scan A f426a01a0383
nono AGENTS.md is an instructions file published in the GitHub repository nolabs-ai/nono (4,686 stars, last pushed yesterday), licensed Apache-2.0. It adds 945 tokens to every session, about $0.0038 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.