vinkius-deploy
How to deploy MCP servers to Vinkius Edge using `vurb deploy`. Use this skill whenever deploying, configuring remote settings, preparing an entrypoint for edge deployment, or troubleshooting deploy-related errors. Activate when the user says "deploy", "publish to edge", "vurb deploy", "push to Vinkius", mentions .vurbrc, VURB_DEPLOY_TOKEN, or any edge deployment topic.
Best use case
vinkius-deploy is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
How to deploy MCP servers to Vinkius Edge using `vurb deploy`. Use this skill whenever deploying, configuring remote settings, preparing an entrypoint for edge deployment, or troubleshooting deploy-related errors. Activate when the user says "deploy", "publish to edge", "vurb deploy", "push to Vinkius", mentions .vurbrc, VURB_DEPLOY_TOKEN, or any edge deployment topic.
Teams using vinkius-deploy should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/vinkius-deploy/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How vinkius-deploy Compares
| Feature / Agent | vinkius-deploy | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
How to deploy MCP servers to Vinkius Edge using `vurb deploy`. Use this skill whenever deploying, configuring remote settings, preparing an entrypoint for edge deployment, or troubleshooting deploy-related errors. Activate when the user says "deploy", "publish to edge", "vurb deploy", "push to Vinkius", mentions .vurbrc, VURB_DEPLOY_TOKEN, or any edge deployment topic.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Deploying to Vinkius Edge
`vurb deploy` bundles your MCP server into a self-contained fat bundle, compresses it, and uploads it to Vinkius Edge where it runs inside a V8 Isolate.
## Quick Start
```bash
# 1. Configure the remote (once per project)
vurb remote --server-id <uuid-from-dashboard>
# 2. Set the deploy token in .env
echo "VURB_DEPLOY_TOKEN=<connection-token-from-dashboard>" >> .env
# 3. Deploy
vurb deploy
```
> The server UUID and connection token are obtained from the Vinkius Cloud dashboard after creating a server instance.
## Configuration Files
### `.vurbrc` (auto-generated by `vurb remote`)
```json
{
"remote": "https://cloud.vinkius.com",
"serverId": "abc-123-def"
}
```
- Auto-added to `.gitignore`
- `remote` defaults to `https://cloud.vinkius.com` (omit to use default)
- `serverId` is the target server UUID from the dashboard
### `.env` (deploy token)
```env
VURB_DEPLOY_TOKEN=your-connection-token-here
```
The token authenticates the deploy request. It can also be passed via `--token <value>` flag.
## Edge Constraints — CRITICAL
V8 Isolates do **NOT** have filesystem or process access. The following patterns are incompatible with edge deployment:
### ❌ `autoDiscover()` — Requires `fs.readdir`
```typescript
// WRONG — autoDiscover() scans the filesystem at runtime
await autoDiscover(registry, fileURLToPath(new URL('./tools', import.meta.url)));
// CORRECT — explicit imports for edge
import { listUsers } from './tools/users.js';
import { getInvoice } from './tools/billing.js';
registry.register(listUsers);
registry.register(getInvoice);
```
### ❌ `SandboxEngine` — Requires `child_process` and `fs`
SandboxEngine cannot run inside V8 isolates. Remove it for edge deploys — tool code runs directly in the isolate.
### ❌ `Inspector` / `@vurb/inspector` — Requires Node.js IPC
The TUI inspector cannot run inside V8 isolates.
### ⚠️ `fast-redact` — Uses `Function` constructor
May have limited support. Verify it works in your isolate before deploying.
## Edge-Compatible Server Entrypoint
See [references/example-edge-server.ts](references/example-edge-server.ts) for a complete example.
Key differences from a standard `server.ts`:
```diff
- import { autoDiscover } from '@vurb/core';
+ // Explicit imports — no filesystem in V8 isolates
+ import { listProducts } from './tools/product.js';
+ import { getInvoice } from './tools/billing.js';
const registry = f.registry();
- await autoDiscover(registry, ...);
+ registry.register(listProducts);
+ registry.register(getInvoice);
```
## Deploy Pipeline — What Happens Under the Hood
```
read .vurbrc + .env
↓
resolve entrypoint (src/server.ts → src/index.ts → server.ts → index.ts)
↓
check for edge-incompatible APIs (autoDiscover, SandboxEngine, Inspector)
↓
esbuild bundle (IIFE, platform: browser, es2022, tree-shaking, minify)
↓
size check (max 500KB raw)
↓
gzip compress + SHA256 hash
↓
POST /servers/:serverId/deploy (Bearer token)
↓
response: { deployment_id, server_name, url }
```
### Bundle Characteristics
| Property | Value |
|---|---|
| Format | IIFE (Immediately Invoked Function Expression) |
| Platform | `browser` (no Node.js APIs) |
| Target | ES2022 |
| Max bundle size | 500KB (raw, before gzip) |
| Dependencies | All bundled inside (zod, vurb, MCP SDK) |
| Node.js modules | Aliased to edge stubs (AST-compatible, never called) |
## CLI Reference
### `vurb deploy`
Bundle, compress, and deploy to Edge.
```bash
vurb deploy # auto-detect entrypoint
vurb deploy --server ./src/app.ts # explicit entrypoint
vurb deploy --token <token> # override VURB_DEPLOY_TOKEN
vurb deploy --allow-insecure # suppress HTTP plaintext warning
```
### `vurb remote`
Configure the target server and API endpoint.
```bash
vurb remote # show current config
vurb remote --server-id <uuid> # set server ID (uses default cloud)
vurb remote https://custom.api.com # override API endpoint
vurb remote https://custom.api.com --server-id <uuid> # both at once
```
### `vurb dev`
Start HMR dev server with auto-reload (for local development).
```bash
vurb dev --server ./src/server.ts
vurb dev --server ./src/server.ts --dir ./src/tools
```
### `vurb lock`
Generate or validate capability lockfile.
```bash
vurb lock # generate/update lockfile
vurb lock --check # verify (CI gate)
```
### `vurb create`
Scaffold a new Vurb server project.
```bash
vurb create my-server
vurb create my-server -y # skip prompts
vurb create my-server --vector prisma --transport sse
```
### `vurb inspect`
Launch the real-time TUI dashboard (requires `@vurb/inspector`).
```bash
vurb inspect --demo # built-in simulator
vurb insp --pid 12345 # connect to server PID
```
## Troubleshooting
| Error | Cause | Fix |
|---|---|---|
| `run: vurb remote --server-id <uuid>` | No `.vurbrc` or missing `serverId` | Run `vurb remote --server-id <uuid>` |
| `set VURB_DEPLOY_TOKEN=<token> in .env` | Token not found | Add token to `.env` or pass `--token` |
| `connection token revoked or invalid` (401) | Invalid or expired token | Generate a new token in the dashboard |
| `connection token does not belong to this server` (403) | Token/server mismatch | Check your server-id matches the token |
| `server not found` (404) | Invalid server UUID | Verify the server exists in the dashboard |
| `bundle too large: NNNkb (max 500KB)` | Fat bundle exceeds limit | Remove unused imports, check tree-shaking |
| `autoDiscover() uses fs.readdir` | Edge-incompatible API | Switch to explicit `registry.register()` |
| `DNS resolution failed` | Wrong remote URL | Check `vurb remote` config |
| `connection refused` | API not running | Verify the remote URL is reachable |
## Entrypoint Auto-Detection
When `--server` is not provided, the CLI probes these paths in order:
1. `src/server.ts`
2. `src/index.ts`
3. `src/server.js`
4. `src/index.js`
5. `server.ts`
6. `index.ts`
7. `server.js`
8. `index.js`Related Skills
vurb-ts-development
How to build production MCP servers with Vurb.ts using the MVA (Model-View-Agent) pattern. Use this skill whenever writing, modifying, or reviewing Vurb.ts code — including tools, Presenters, Models, middleware, prompts, routers, tests, or server configuration. Activate even when the user just says "create a tool", "add an endpoint", "write a Presenter", or mentions @vurb/core, defineModel, definePresenter, initVurb, FluentToolBuilder, or any Vurb.ts API. This skill covers the entire framework surface.
deployment-patterns
Deployment workflows, CI/CD pipeline patterns, Docker containerization, health checks, rollback strategies, and production readiness checklists for web applications. Use when setting up deployment infrastructure or planning releases.
makepad-deployment
CRITICAL: Use for Makepad packaging and deployment. Triggers on: deploy, package, APK, IPA, 打包, 部署, cargo-packager, cargo-makepad, WASM, Android, iOS, distribution, installer, .deb, .dmg, .nsis, GitHub Actions, CI, action, marketplace
expo-deployment
Deploy Expo apps to production
devops-deploy
DevOps e deploy de aplicacoes — Docker, CI/CD com GitHub Actions, AWS Lambda, SAM, Terraform, infraestrutura como codigo e monitoramento.
deployment-validation-config-validate
You are a configuration management expert specializing in validating, testing, and ensuring the correctness of application configurations. Create comprehensive validation schemas, implement configurat
deployment-procedures
Production deployment principles and decision-making. Safe deployment workflows, rollback strategies, and verification. Teaches thinking, not scripts.
deployment-pipeline-design
Architecture patterns for multi-stage CI/CD pipelines with approval gates and deployment strategies.
deployment-engineer
Expert deployment engineer specializing in modern CI/CD pipelines, GitOps workflows, and advanced deployment automation.
kubernetes-deployment
Kubernetes deployment workflow for container orchestration, Helm charts, service mesh, and production-ready K8s configurations.
modal-deployment
Run Python code in the cloud with serverless containers, GPUs, and autoscaling using Modal. This skill enables agents to generate code for deploying ML models, running batch jobs, serving APIs, and scaling compute-intensive workloads.
vercel-github-actions-deploy
Set up GitHub Actions to deploy any Vercel project using the Git Author Override method, enabling teammates to deploy on the free Hobby plan. Use when the user asks about Vercel deployment via GitHub Actions, CI/CD for Vercel, letting teammates deploy on Vercel free plan, bypassing Vercel's Hobby plan deploy restrictions, or automating Vercel production deploys. Covers workflow setup, GitHub Secrets configuration, and package manager variants (bun, npm, pnpm).