Packages

AI-optimized web search API designed for LLM agents that returns clean, pre-extracted textual content from search results rather than raw HTML or SERP metadata.

Think MCP server providing AI agents with a dedicated thinking and reflection tool — enabling agents to pause and reason through problems, maintain scratchpad-style working memory during complex tasks, structure multi-step reasoning chains, and separate thinking from action execution, implementing Anthropic's extended thinking pattern as an explicit MCP tool.

Provides a cloud platform for web scraping and browser automation via a marketplace of 1,500+ ready-made Actors (containerized scrapers) plus a REST API and SDK for running, managing, and scheduling them.

Provides cloud-hosted browser infrastructure for AI agents, offering managed Chromium sessions accessible via Playwright or Puppeteer with session recording, proxies, and CAPTCHA solving.

Neural web search API purpose-built for LLM agents. Exa uses embeddings to search the web semantically — understanding concepts and relationships, not just keyword matching. Returns full page content (not just snippets) in clean Markdown format, enabling agents to actually read and process search results. Also provides FindSimilar (find URLs similar to a given URL), GetContents (fetch clean content from any URL), and neural search for specific content types (news, papers, tweets, GitHub).

Official JetBrains MCP Server Plugin enabling AI agents to interact with JetBrains IDEs (IntelliJ IDEA, PyCharm, WebStorm, etc.) — reading and editing files, running code, accessing project structure, executing terminal commands, and integrating JetBrains IDE capabilities into agent-driven development workflows.

Neon serverless PostgreSQL API — fully managed serverless Postgres with instant database branching, enabling agents to create isolated database branches for dev/test/staging, scale-to-zero compute, and standard PostgreSQL queries via standard connection strings.

Official Okta MCP server enabling AI agents to interact with Okta's identity platform — managing users, groups, applications, and policies for workforce and customer identity management.

Official Square MCP server enabling AI agents to interact with Square's commerce platform — managing catalog items, querying orders and payments, managing customers and loyalty programs, handling inventory, and integrating Square's in-person and online commerce APIs.

Official Stripe MCP server from the Stripe agent toolkit. Enables AI agents to interact with Stripe's payment APIs — creating payment intents, managing customers, subscriptions, invoices, products, prices, and refunds. Exposes core Stripe financial operations to LLM agents.

Official Browserbase MCP server enabling AI agents to control cloud browsers — navigating web pages, clicking elements, filling forms, taking screenshots, extracting web content, and automating browser interactions through Browserbase's managed headless browser infrastructure.

Official Cloudinary MCP server enabling AI agents to manage media assets — uploading images and videos, applying transformations, searching the media library, and generating optimized delivery URLs.

Industry-leading text-to-speech API with 1000+ voices, voice cloning, and ultra-low-latency streaming audio generation for voice AI applications.

Official LaunchDarkly MCP server enabling AI agents to interact with LaunchDarkly's feature management platform — reading flag status, toggling flags, managing targeting rules, and monitoring experiments.

Official ChromaDB MCP server enabling AI agents to interact with ChromaDB — the open-source AI-native vector database — storing and querying document embeddings, managing collections, performing semantic similarity search, and integrating ChromaDB's vector storage into agent-driven RAG pipelines, semantic memory, and knowledge retrieval workflows.

Official ClickHouse MCP server enabling AI agents to query ClickHouse databases — running SQL analytics, exploring table schemas, executing OLAP queries, and accessing ClickHouse's high-performance columnar storage for real-time analytics at scale.

Provides a reactive TypeScript-native database where queries automatically re-run and push updates to subscribers whenever underlying data changes, combining backend logic, storage, and real-time sync in one platform.

Python Interpreter MCP server from Sentry enabling AI agents to execute Python code — running arbitrary Python scripts, accessing installed packages, maintaining execution state across calls, performing data analysis, and integrating interactive Python code execution into agent-driven development, data science, and automation workflows.

Rust Docs MCP server enabling AI agents to search and retrieve Rust documentation — accessing the Rust standard library docs, querying crate documentation on docs.rs, searching for Rust functions and types, and integrating Rust's official documentation into agent-driven Rust development assistance and code generation workflows.

Real-time AI security layer that detects and blocks prompt injection attacks, jailbreak attempts, and harmful LLM outputs with sub-50ms latency.