Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
@layervai/qurl-mcp
⚠️ Renamed from `@layerv/qurl-mcp` in v0.4.0. The old package is deprecated and will not receive further updates. If you're using @layerv/[email protected], swap the scope in your MCP client config — same binary, same API key, no other changes.MCP server for qURL™ secure link management.
Quantum URL (qURL) · The internet has a hidden layer. This is how you enter.
What it does
qURL MCP Server is a Model Context Protocol server that lets AI agents (Claude, GPT, Cursor, etc.) create, resolve, list, and manage qURL secure links natively. It connects to the qURL API over stdio transport, so any MCP-compatible client can use it without custom integration code.
Quick Start
Add the server to your MCP client configuration (Claude Desktop, Claude Code, etc.):
{
"mcpServers": {
"qurl": {
"command": "npx",
"args": ["@layervai/qurl-mcp"],
"env": { "QURL_API_KEY": "lv_live_xxx" }
}
}
}Replace lv_live_xxx with your actual API key. The key must have the appropriate scopes for the tools you intend to use (see below).
Available Tools
| Tool | Description | Required Scope |
|---|---|---|
create_qurl | Create a secure, policy-bound link to a protected resource | qurl:write |
resolve_qurl | Resolve an access token to get the target URL and grant network access | qurl:resolve |
list_qurls | List active qURLs with optional pagination | qurl:read |
get_qurl | Get details of a specific qURL by resource ID | qurl:read |
delete_qurl | Revoke a qURL, immediately invalidating the link | qurl:write |
extend_qurl | Extend the expiration of an active qURL (alias for update_qurl) | qurl:write |
update_qurl | Update expiration, tags, or description on an active qURL | qurl:write |
mint_link | Mint a new access link for an existing protected resource | qurl:write |
batch_create_qurls | Create multiple qURLs in a single call | qurl:write |
revoke_qurl_token | Revoke one qURL token without revoking sibling tokens on the resource | qurl:write |
update_qurl_token | Update expiry, label, policy, or session limits on one qURL token | qurl:write |
list_qurl_sessions | List active access sessions for a qURL resource | qurl:read |
terminate_qurl_sessions | Terminate one or all active sessions for a qURL resource | qurl:write |
Available Resources
| URI | Name | Description |
|---|---|---|
qurl://links | Active qURL Links | List of all active qURL links |
qurl://usage | qURL Usage & Quota | Current quota and usage information |
Configuration
| Environment Variable | Required | Description | Default |
|---|---|---|---|
QURL_API_KEY | Conditional (see description) | API key with appropriate scopes (qurl:read, qurl:write, qurl:resolve). The server boots without it so MCP introspection (tools/list, resources/list, prompts/list) works for directory probes — required only on the first tool call or resource read, where invocations surface a typed missing_api_key error until the key is set. | -- |
QURL_API_URL | No | qURL API base URL | https://api.layerv.ai |
Docker
A multi-stage Dockerfile is included for container-based deployment:
docker build -t qurl-mcp .
docker run -i -e QURL_API_KEY=lv_live_xxx qurl-mcpThe image runs as the non-root node user, ships only production dependencies, and uses tini as PID 1 for clean signal handling.
If a tool call returns missing_api_key despite QURL_API_KEY looking set, check stderr for the boot-time warning — some MCP hosts hide stderr, and the warning is the fastest way to spot a whitespace-only or unset value:
docker logs <container> # if running detached
docker run -i -e QURL_API_KEY=lv_live_xxx qurl-mcp 2>&1 # interactiveDevelopment
npm install
npm run build
npm test
npm run lintAdditional commands:
npm run dev # Watch mode (rebuild on changes)
npm run format # Format source with Prettier
npm run format:check # Check formatting without modifying filesLicense
MIT -- LayerV AI