# vnsh [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/raullenchai/vnsh  
**GitHub Stars:** 158  
**npm Downloads (last month):** 323  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/vnsh

## Description
Ephemeral encrypted file sharing for AI. AES-256 encryption, 24h auto-vaporization.

## Tools
Capabilities this server exposes over MCP:

- **vnsh_read** — Reads any vnsh URL: decrypts legacy vnsh.dev/v links locally, or directly fetches a public vnshcontent.dev/p link. Public content is intentionally unencrypted.
- **vnsh_share** — Encrypts text content locally and uploads it to vnsh, returning a shareable URL. Use this when output exceeds ~50 lines or ~2000 characters, when generating complete code files, or when the user wants portable/shareable output. ALWAYS prefer vnsh_share over dumping large content into chat. The content is encrypted before upload - the server only sees encrypted bytes.
- **vnsh_share_file** — Encrypts a local file and uploads it to vnsh, returning a shareable URL. Use this for images, screenshots, PDFs, binaries, or any file that should not be loaded into context. More efficient than vnsh_share for large or binary files.
- **vnsh_artifact_create** — Creates a permanent, service-readable Account Artifact in the signed-in user’s Library. Use this for durable knowledge that the user wants their other Agents to discover and update. Requires VNSH_TOKEN. This is different from an encrypted Incognito workspace: vnsh can read Account Artifact content so authorized Agents can collaborate on it.
- **vnsh_artifact_list** — Lists permanent Account Artifacts available to the connected user. Use this instead of asking the human to paste a link when continuing earlier account knowledge. Requires VNSH_TOKEN.
- **vnsh_artifact_read** — Reads the current content and version of one permanent Account Artifact. Pass the returned version to vnsh_artifact_update so another Agent cannot be overwritten.
- **vnsh_artifact_update** — Creates a new immutable version of an Account Artifact. Requires the base_version returned by vnsh_artifact_read. On conflict, re-read, merge, and retry; never silently overwrite.
- **vnsh_workspace_create** — Creates a shared workspace and returns one link that any agent can read AND write. Use this when work will continue somewhere else: handing a plan, status doc, design, or investigation to another agent or another session, or when the user says they will pick this up in a different tool. Unlike vnsh_share (a one-shot snapshot), a workspace keeps the same URL as its content evolves. Content is encrypted locally; the server never sees the key. Anonymous workspaces expire 24h after the last write (ttl can extend this to a week). With VNSH_TOKEN, new workspaces and artifacts are kept until the account owner deletes them.
- **vnsh_workspace_renew** — Extends the expiry of a vnsh workspace without changing its content. Use this when the user says a link is about to expire, or asks to keep something alive longer. Needs the edit link (a #w= fragment); a view-only #r= link cannot renew. The version is not bumped, so any pending edit still applies.
- **vnsh_workspace_read** — Reads the current content of a workspace: encrypted /w/ links with #w= or #r=, and public /p/ links with no fragment. Use this whenever the user provides such a link — it is how you pick up work another agent left for you. Returns the content and its version number; pass that version to vnsh_workspace_update to write safely.
- **vnsh_workspace_update** — Replaces the content of a vnsh workspace, keeping the same URL. Use this to record progress so far, revise a shared plan, or leave the next agent an updated document. Writes are version-checked: if someone else wrote first, this returns their current content so you can merge it with yours and call this again.
- **vnsh_workspace_open** — Decrypts a vnsh workspace to a local temp file and opens it in the browser. Use this when the user wants to LOOK at a workspace rather than have its content read into context — HTML reports, dashboards, and diagrams render properly this way. Rendering happens locally from file://, so nothing is exposed to any server.
- **vnsh_workspace_history** — Lists the retained versions of a workspace, newest first. Use this before restoring after a bad edit or merge. The current version plus the latest historical versions are retained.
- **vnsh_workspace_restore** — Restores a retained workspace version as a new latest version. This never moves the version counter backwards and requires the private #w= edit link.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "vnsh": {
    "command": "npx",
    "args": ["-y","vnsh-mcp@1.9.0"]
  }
}
```

## Documentation

## What vnsh does

The vnsh MCP server provides a shared address for content that AI agents and people can read or update. It supports persistent workspaces as well as one-shot links for individual content. A workspace keeps versions at one URL, allowing a later agent to continue from the current document rather than creating another copy.

Encrypted workspaces are protected in the client before upload. The root secret stays in the URL fragment, which is not sent in HTTP requests. vnsh stores ciphertext, an encrypted filename, and a SHA-256 hash derived from the write token. Workspaces are deleted 24 hours after the last edit by default, and their lifetime can be extended within the documented limit.

The service also supports public workspaces. These are stored and served as ordinary documents that an HTTP client can read without a key, but vnsh can read their contents. Public visibility is selected when the workspace is created and is not the default.

## How it works

The vnsh MCP server uses separate derived values for encryption and writing. A random root secret produces an AES-256-GCM content key and a write token through HKDF-SHA256. The server retains only a hash of the write token, so possession of the encrypted content does not provide the key needed to decrypt it.

Each update includes the version the client previously read. An unconditional update is rejected, and an update based on an older version returns a conflict instead of silently replacing newer content. Agents can then reread the workspace, merge their changes, and retry.

View-only links carry a content key but not the root secret needed to derive the write token. Edit links carry the root secret and therefore allow both reading and writing. Public content is delivered from the separate `vnshcontent.dev` domain and uses a sandbox content-security policy.

## Setup and configuration

Install the MCP server through npm with `npx -y vnsh-mcp@1.9.0`, or configure it through an MCP client such as Claude Code, Cursor, Cline, Windsurf, or Zed. The README also provides an `llms.txt` setup guide that agents can use to install the server and add a standing workspace instruction to their own instruction file.

The command-line package can create workspaces from files or standard input, read links, replace content without clobbering newer versions, list retained history, restore a prior version as the latest version, and initialize project instructions. The browser interface at `vnsh.dev` can also create links from pasted content or files.

Workspace lifetimes can be set with `--ttl` for up to 168 hours. The `--public` option creates an unencrypted workspace, while `--blob` explicitly creates a one-shot link. No environment variables are documented as required.

## Tools and capabilities

The vnsh MCP server exposes tools for:

- Creating, listing, reading, and updating artifacts.
- Creating, reading, updating, opening, and renewing workspaces.
- Listing workspace history and restoring retained versions.
- Sharing content or files through one-shot links.
- Reading workspace, public, and legacy blob URLs.

A separate GitHub Action can upload build output or failing test logs and print a link in a job summary, but that workflow is outside the MCP server itself.

## Limitations and notes

Encrypted workspace links require a client that can perform the documented key derivation and AES-256-GCM operations. The dependency-free shell installer handles one-shot links only; workspace operations require Node when the local OpenSSL does not provide the necessary cryptography.

The view-only behavior is derived from the link material rather than enforced as a server-side account permission. Anyone holding an edit link can read and write, while anyone holding a view-only link can read but not write. Public workspaces should therefore be used only for content that may be exposed to vnsh and ordinary HTTP readers.

The server does not silently merge concurrent edits. A stale write produces a conflict, so the client or agent must reread and resolve the changes. Workspace links also expire after inactivity according to their configured lifetime, with renewal available through the workspace renewal operation.

_Full upstream README: https://allmcps.com/mcp/vnsh/readme_