Skip to content

Arcade

Arcade is an AI tool-calling platform and Model Context Protocol (MCP) runtime that provides a secure environment for AI agents to interact with third-party software and enterprise services.

Paid · Web, API

stork
I am a painting.
One still painting and one typed sentence.
This painting was never filmed.
A photo, a cartoon, a painting or a 3D character. From 10 cents a minute. Free to try.
Make a photo talkDevelopers: there is an API

What is Arcade?

Arcade is an AI tool-calling platform and Model Context Protocol (MCP) runtime developed by Arcade.dev that enables AI agents to securely and reliably interact with enterprise and consumer services. It provides a runtime environment for AI agents, facilitating connections with third-party software, managing authentication, tool access, and governance.

Key Features of Arcade

Arcade provides a comprehensive suite of features designed to enable secure, reliable, and governed AI agent interactions with external services. Its architecture focuses on robust authentication, dynamic permissions, and an extensible tool ecosystem.

  • AI tool-calling platform and runtime environment for agents.
  • Manages authentication, tool access, and governance for AI agents.
  • Open-source marketplace with community-managed connectors.
  • SDK for developing custom tools and Model Context Protocol (MCP) servers.
  • Agent authorization with dynamic permissions and Identity Provider (IDP) integration.
  • Agent-optimized tools designed for reliable API calls, preventing hallucinations and silent failures.
  • Central control plane for policy enforcement and auditing of agent actions.
  • SOC 2 compliant infrastructure with Single Sign-On (SSO), Role-Based Access Control (RBAC), and full audit logs.
  • Flexible deployment options including Cloud, on-premise, air-gapped, and hybrid environments.
  • Integrates with any Large Language Model (LLM), framework, identity provider, and MCP client.

Who Should Use Arcade?

Arcade is primarily utilized by developers and enterprises seeking to deploy production-ready AI agents that require secure, authenticated access to a wide array of third-party and internal systems. Its capabilities address critical challenges in agent deployment, security, and compliance.

  • Automating Workflows: Building production-ready AI agents to manage support tickets, schedule meetings, send emails, update CRM records, and coordinate cross-system workflows.
  • Secure Enterprise Integration: Enabling AI agents to securely access and act within enterprise systems in regulated sectors like financial services and healthcare.
  • Multi-User Authorization: Solving the challenge of agents acting across multiple user contexts by securely managing individual user permissions via OAuth 2.0 and API keys.
  • Developer Enablement: Providing developers with pre-built "agent tools" and an SDK for customized integration with any API, data, logic, or system.
  • Ensuring Compliance and Auditability: Deploying agents with a central control plane for policy enforcement, full audit logs, and SOC 2 compliance.

How to Use Arcade

To utilize Arcade, developers typically integrate its SDK into their AI agent frameworks or leverage its pre-built tools and connectors. The platform provides a secure runtime for agent actions, managing authentication and authorization.

  1. Access the Arcade platform via arcade.dev and explore the pre-built catalog of agent tools.
  2. Utilize the Arcade SDK to develop custom tools or Model Context Protocol (MCP) servers for specific API integrations.
  3. Configure agent authorization by integrating with existing Identity Providers (IDPs) and setting dynamic permissions.
  4. Deploy AI agents within the Arcade runtime environment, ensuring secure and governed interactions with third-party services.
  5. Monitor agent behavior and actions using the platform's central control plane for policy enforcement and auditing.
  6. Leverage the open-source marketplace to find or contribute community-managed connectors for various API endpoints.

Arcade Pricing & Plans

Arcade operates on a paid model, offering a free tier for initial exploration and development. Specific pricing details for advanced features and enterprise plans are available on the official Arcade pricing page. The platform's pricing structure is designed to support various scales of AI agent deployment, from individual developers to large enterprises requiring robust security and compliance features.

  • Free Tier: Available for initial use and evaluation.
  • Paid Plans: Specific pricing for advanced features, enterprise-grade security, and compliance are detailed on arcade.dev/pricing.

Pros

  • Provides a secure action layer for AI agents, managing OAuth 2.0, API keys, and user tokens without exposing credentials to the LLM.
  • Offers a comprehensive authorization system with granular permissions for multi-user agent contexts.
  • Includes an open-source marketplace with community-managed connectors and an SDK for custom tool development.
  • SOC 2 compliant with features like SSO, RBAC, and full audit logs, addressing enterprise security requirements.
  • Accelerates AI agent deployment by handling complex authentication, tool integration, and governance infrastructure.
  • Received $60 million in Series A funding in June 2026, indicating strong investor confidence and growth potential.

Cons

  • Pricing for advanced features may be high for small teams or startups, as noted in user feedback.
  • Some users have mentioned limitations in customization options for certain features.
  • Analytics capabilities have been cited as an area for improvement by users.
  • A learning curve may exist for mastering best practices and fully leveraging the platform's capabilities.

Arcade vs Competitors

Arcade positions itself as a foundational technology for the "agentic era," specifically as the industry's first authenticated tool-calling platform for Agentic AI and an MCP runtime. Its core differentiator is a secure action layer for AI agents, built by former Okta identity engineers, which handles OAuth and manages user tokens, API keys, and secrets without exposing credentials to the AI model. This just-in-time authorization model significantly reduces security risks compared to alternatives.

Arcade

Paid · Web, API

Enjoying this? Get one like it in your inbox each morning.

one email a day · unsubscribe in two clicks · no third-party tracking