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/raullenchai/vnsh/claude-mdgit clone --depth 1 https://github.com/raullenchai/vnshWrote 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/raullenchai/vnsh/claude-md)<a href="https://agentmods.dev/instructions/raullenchai/vnsh/claude-md"><img src="https://agentmods.dev/badge/instructions/raullenchai/vnsh/claude-md.svg" alt="Measured on agentmods" 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 | $0.00732 | $0.00732 |
| Opus 5 | $0.00366 | $0.00366 |
| Sonnet 5 | $0.00146 | $0.00146 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
vnsh CLAUDE.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 6d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vnsh contributor context
vnsh is an ephemeral shared workspace for people and AI agents. Its design principle is server-side blindness: encrypted workspace plaintext and keys never reach the service. The repository is MIT licensed.
Current product model
There are three link types:
/w/{id}#w={secret}is an encrypted read/write workspace./w/{id}#r={key}is an encrypted, cryptographically view-only workspace.https://vnshcontent.dev/p/{id}is an explicitly public workspace.
Encrypted workspaces use a random 32-byte root secret. HKDF-SHA256 derives an
AES-256-GCM content key and a separate write token. The service stores ciphertext
and SHA-256(write token), never either secret. Updates require If-Match, so a
stale client receives 412 rather than silently overwriting a newer version.
The legacy one-shot /v/ blob format remains supported and uses AES-256-CBC with
its key and IV in the URL fragment. Do not describe CBC as the workspace format.
Workspaces expire 24 hours after the last successful write by default. A caller may request a supported TTL up to seven days. R2 object data and custom metadata are the source of truth; KV is no longer part of the runtime architecture.
Repository
worker/: Cloudflare Worker, R2 API, browser UI, public-content host, cleanupcli/npm/: Node CLI (vnandvnsh)mcp/: MCP server for workspace create/read/update/open and legacy share/readextension/: Chrome extension for previews and capture bundlesscripts/smoke.mjs: production end-to-end verification
The Worker serves vnsh.dev. Explicitly public documents are isolated on the
separate registrable domain vnshcontent.dev. The content host is dispatched
before all normal routes and serves only its narrow public surface. The legacy
vnsh.dev/p/ compatibility window is controlled by LEGACY_PUBLIC_UNTIL; after
the cutoff it intentionally returns 410 rather than redirecting.
Development and deployment
From the repository root:
npm test
npm run dev
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.
- 6d ago First seen · 81 lines · 732 tokens per session scan A 9a2df5024b5b
vnsh CLAUDE.md is an instructions file published in the GitHub repository raullenchai/vnsh (156 stars, last pushed 22d ago), licensed MIT. It adds 732 tokens to every session, about $0.0037 per session on Opus 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
start-technologies AGENTS.md
AGENTS.md instructions for Start9Labs/start-technologies, covering agents.md, layout, build & test (run from the repo root), branches and channels and releases.
Compartment GEMINI.md
Gemini CLI instructions for MaxFreedomPollard/Compartment, a project described as: Encrypted, fully offline agentic memory. One click install, GUI w/ memory map, all OS and agents. Superior memory creation, storage and retrieval.
awesome-trustworthy-deep-learning AGENTS.md
Instructions for MinghuiChen43/awesome-trustworthy-deep-learning, covering agents.md, project context, default task: add an arxiv paper, 1. normalize the arxiv id and 2. fetch metadata reproducibly.
pasteguard AGENTS.md
Instructions for sgasser/pasteguard, covering pasteguard agent instructions, project, stack, commands and workflow.
presidio-research AGENTS.md
AGENTS.md instructions for data-privacy-stack/presidio-research, covering agents.md — codebase patterns, project setup, canonicalmapper, five issue types (issuetype enum) and resolution dataclass fields.
app-privacy-policy-generator AGENTS.md
AGENTS.md instructions for nisrulz/app-privacy-policy-generator, covering agents.md, project overview, source layout, build and code conventions.