Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/ripmilla/netwalk/agents-mdgit clone --depth 1 https://github.com/ripmilla/netwalkWhat 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 | $0.16337 | $0.16337 |
| Opus 5 | $0.08169 | $0.08169 |
| Sonnet 5 | $0.03267 | $0.03267 |
| Haiku 4.5 | $0.01634 | $0.01634 |
Grade C, and why
netwalk AGENTS.md scanned grade C with 1 finding 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 3d 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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
A **path** to a key file (`~/.ssh/id_ed25519`, `C:\Users\me\.ssh\id_rsa`) is not a secret and is How it starts
The opening of the file, as written. The whole thing — 1,152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
netwalk — instructions for AI coding agents
netwalk is a read-only network survey toolkit. This file is the agent-agnostic form of the
same instructions that ship as Claude Code skills under skills/. Any agent that reads a project
instruction file (AGENTS.md, GEMINI.md, .cursor/rules, .clinerules, CONVENTIONS.md …)
can work from this.
Replace <TOOLKIT> below with wherever you cloned this repository.
The three promises, and why they are not optional
1. netwalk never changes a device. Every command is checked against a per-vendor read-only
allowlist in scripts/netwalk_policy.py before it is sent. Config writes, counter clears,
service restarts, reboots and command smuggling are refused by the tool, not by your good
intentions. Never SSH to a surveyed device directly — always go through
scripts/netwalk_exec.py, so the guarantee holds and the command lands in the evidence log.
2. Credentials never enter the conversation. They are typed into a page served on the user's
own 127.0.0.1 by scripts/netwalk_cred.py, stored in a private file, and read by the exec
wrapper — never by you. Do not ask for a password in chat, do not accept one if offered, and never
open ~/.netwalk/creds/*.json. You do not need the value; you pass a site slug and a host id.
3. netwalk never sweeps a range nobody authorised. The address sweep in
scripts/netwalk_sweep.py refuses any range that is not written into the site's scope.json with
the name of the person who authorised it. A supernet of an authorised range is refused, public
address space needs a second explicit flag, anything larger than a /16 is refused outright, and
there is no --force. Crawling from a device the user gave you a credential for is one thing;
probing addresses nobody named is another, and in several places netwalk gets used it is an offence.
If a read-only command is wrongly blocked, add it to the allowlist, run
python3 <TOOLKIT>/tests/test_policy.py, and say that you did. Never route around the wrapper.
The same rule applies to the sweep gate: get the authorisation, do not edit the scope file yourself.
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.
- 3d ago First seen · 1,152 lines · 16,337 tokens per session scan C 78cf898d15f0
netwalk AGENTS.md is an instructions file published in the GitHub repository ripmilla/netwalk (75 stars, last pushed 10d ago), licensed MIT. It adds 16,337 tokens to every session, about $0.0817 per session on Opus 5. A static security scan graded it C with 1 finding (reaches for credential files). 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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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 oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.