Get Started Free

Discover

AI Stack

Code Execution & Tooling for AI Agents

E2B. Modal. Jupyter. Docker. Replit. Code, queries, and containers - connected, governed, and production-ready.

Curated by the Vinkius team — 5 production-ready MCP servers reviewed, tested, and ready to connect to your AI agents. Create your free account and start connecting in seconds — no infrastructure to manage, no code to write. Just plug in and let your agents work.

Ask AI about these MCP Servers

Open in ChatGPT Open in Claude Open in Perplexity

E2B MCP Server

01MCP Server

E2B MCP Server

The #1 code sandbox for AI agents - isolated execution in under 150ms.

E2B is the gold standard for AI code execution. This MCP Server gives your agent a fully isolated cloud sandbox that boots in under 150ms - run Python, JavaScript, Bash, or any language without touching your infrastructure. Each execution gets its own microVM with filesystem, network, and process isolation. Used by thousands of AI companies for safe, deterministic code execution. Your agent writes code, E2B runs it - zero risk.

Isolated microVM code execution

Multi-language support (Python, JS, Bash)

Sub-150ms sandbox boot time

Connect your agent

Browserbase

Browserbase

Cloud browser sandbox with CDP control

Hyperbrowser

Hyperbrowser

Headless browser infrastructure for AI agents

Browserless

Browserless

Headless Chrome execution as a service

Playwright Cloud

Playwright Cloud

Remote Playwright browser automation

Google BigQuery MCP Server

02MCP Server

Google BigQuery MCP Server

Petabyte-scale SQL engine - query terabytes in seconds, serverless.

cloud.google.com/bigquery

BigQuery is Google's massively parallel SQL engine that processes petabytes without provisioning infrastructure. This MCP Server gives your agent direct access to Standard SQL execution at warehouse scale - query live datasets, monitor active jobs, and inspect table schemas across your entire data estate. From ad-hoc analytics to real-time pipeline monitoring, your agent runs SQL the way enterprise teams do - fast, governed, and serverless.

Standard SQL execution at petabyte scale

Real-time job monitoring & tracking

Schema inspection across datasets

Connect your agent

Metabase

Metabase

Open-source BI with native SQL querying

Amazon Redshift

Amazon Redshift

Cloud data warehouse with SQL execution

Apache Superset

Apache Superset

SQL Lab & data exploration dashboards

Neon

Neon

Serverless PostgreSQL with branching

Vercel MCP Server

03MCP Server

Vercel MCP Server

The serverless deploy platform - instant global deployment from any AI agent.

Vercel powers the modern web - from Next.js to edge functions, millions of developers deploy here. This MCP Server gives your agent full deployment control: trigger builds, manage environment variables, configure domains, and monitor edge function performance. Your agent goes from code to production in seconds, with automatic preview deployments and global CDN distribution. The fastest path from AI-generated code to live application.

Instant build & deploy from code

Edge function management & monitoring

Preview deployments & domain config

Connect your agent

Heroku

Heroku

The original PaaS - deploy anything instantly

Fly.io

Fly.io

Global container deploy in 35+ regions

Railway

Railway

Deploy containers with zero config

Render

Render

Unified cloud for web services & cron jobs

Cloudflare MCP Server

04MCP Server

Cloudflare MCP Server

Edge compute protecting 20% of the internet - Workers, Pages, and serverless at global scale.

Cloudflare runs on 330+ cities worldwide, protecting and accelerating over 20% of all internet traffic. This MCP Server gives your agent access to Workers serverless compute, Pages deployments, DNS management, and Zero Trust security - all from a single interface. Deploy edge functions in 50ms, manage WAF rules, purge caches, and configure routing. When your agent needs to execute code at the edge of the network, Cloudflare is the infrastructure.

Workers serverless edge compute

Pages deployment & DNS management

Zero Trust security & WAF rules

Connect your agent

GitLab

GitLab

Complete DevOps with built-in CI/CD

CircleCI

CircleCI

CI/CD pipelines trusted by elite teams

Buildkite

Buildkite

Scalable CI/CD with self-hosted runners

Harness

Harness

Enterprise CI/CD with AI-powered pipelines

Modal MCP Server

05MCP Server

Modal MCP Server

Serverless GPU compute - run any code, any model, any scale, instantly.

Modal eliminates infrastructure management for compute-intensive workloads. This MCP Server gives your agent access to serverless GPU and CPU containers that spin up in seconds - no Dockerfiles, no Kubernetes, no provisioning. Run training jobs, batch inference, data pipelines, and custom scripts on A100s or H100s with a single function call. Your agent defines the compute, Modal handles everything else. The fastest way to go from code to running workload.

Serverless GPU containers (A100/H100)

Zero-config compute provisioning

Batch inference & data pipelines

Connect your agent

RunPod

RunPod

GPU pods for AI training & inference

Pipedream

Pipedream

Serverless workflow execution platform

Paperspace

Paperspace

GPU cloud with Jupyter notebooks built-in

Netlify

Netlify

Instant web builds with serverless functions

Code Execution. SQL Queries. Containers. Serverless. Ready for AI Agents.

Create your free account and connect these MCP servers to your AI agents in seconds — no infrastructure to set up, no code to write.

We handle the servers, the security, the updates, and the uptime. You just connect and use.

Try for Free·No credit card