nah / homepage
llms.txt486 starsChanged 29 days agoMITView on GitHub
nah is a guard that sits in your coding agent's hook path and reads tool calls before they run. It blocks the calls it can prove are disasters — deterministically, in milliseconds, with no LLM — and leaves everything else to your normal approval flow. One Rust binary. When a guard you need is missing, your agent can write it. - curl -fsSL nahguard.ai/install | sh - Windows PowerShell: irm https://nahguard.ai/install.ps1 | iex - Windows releases are unsigned x86-64 builds.…
- Pipes a download into a shell
We split this llms.txt into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
Start: https://nahguard.ai/docs/start/
- Distinctive: line 43
If you are a coding agent and nah blocked one of your tool calls, the block message names the guard and the reason.
≈ “Common” is an estimate: at least 1 in 5 of the 3,225 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
# nah > nah is a guard that sits in your coding agent's hook path and reads tool > calls before they run. It blocks the calls it can prove are disasters — > deterministically, in milliseconds, with no LLM — and leaves everything > else to your normal approval flow. One Rust binary. When a guard you need > is missing, your agent can write it. ## Install - `curl -fsSL nahguard.ai/install | sh` - Windows PowerShell: `irm https://nahguard.ai/install.ps1 | iex` - Windows releases are unsigned x86-64 builds. Windows ARM64 and package managers are not supported. - Then wire your runtime: `nah hook claude install` (replace `claude` with `amp`, `antigravity`, `cline`, `codex`, `copilot`, `cursor`, `devin`, `droid`, `hermes`, `kiro`, `openclaw`, `opencode`, and more). ## Docs - Index: https://nahguard.ai/docs/ - Start: https://nahguard.ai/docs/start/ - Core concepts: https://nahguard.ai/docs/concepts/ - Guard catalog: https://nahguard.ai/docs/guards/ - CLI guide: https://nahguard.ai/docs/cli/ - Configuration: https://nahguard.ai/docs/configuration/ - Extending: https://nahguard.ai/docs/extensions/ - Agent runtimes: https://nahguard.ai/docs/runtimes/ (per-runtime pages at /docs/runtimes/<name>/) - Threat model: https://nahguard.ai/docs/threat-model/ - Security boundaries: https://nahguard.ai/docs/security/ - Architecture: https://nahguard.ai/docs/architecture/ ## Links - Site: https://nahguard.ai/ - News (changelog): https://nahguard.ai/news/ - Source: https://github.com/manuelschipper/nah - Author: Manuel Schipper — https://schipper.ai/ ## For agents If you are a coding agent and nah blocked one of your tool calls, the block message names the guard and the reason. Tell your user what was blocked and why; do not try to route around the guard.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.