Features

Metorial provides a library of more than 1000 managed integrations that are verified to work with popular AI agents and ready to use without building connections yourself.

Agent Explorer lets you connect any integration in Metorial and use a real AI model to test it with real data. You can see how an agent will actually use your tools before you put them in production.

Metorial gives each AI agent a unique identity. Every action ties back to a specific agent and the user behind it, and access can be controlled per agent and per session.

Metorial's workspace is a single console for managing everything AI in your organization, agents, integrations, skills, users, and policies. Admins can run it all from one place.

Metorial records every interaction with the platform. Security and compliance teams have a complete, filterable history of what people, AI agents, and machine users did across your integrations and skills.

Metorial deploys your custom MCP servers automatically from GitHub, GitLab, Azure DevOps, and Bitbucket. You ship them using the git workflow you already have.

Connect your own integrations and internal systems to AI agents by building them as custom MCP servers. Metorial hosts, scales, and secures them, with the same access control, tracing, and isolation as every other provider.

Metorial lets you configure how long the data it stores is kept, from session logs and tool calls to audit logs. You can meet compliance requirements with long or short retention as needed.

Deploy any MCP server that can be packaged as a Docker container on Metorial, run it in an isolated enclave, and govern it with the same controls as every other provider.

Enclaves are the fast, secure sandboxes that run every integration on Metorial. Integration code is always isolated and monitored, down to every user, connection, and tool call.

Enterprise search gives you one place to search across your integrations, skills, agents, and logs. You can find information and troubleshoot without hunting through separate views.

Metorial's virtual firewalls govern inbound and outbound network traffic for every integration and MCP server. You can enforce per-workload network policy, block exfiltration, and monitor all traffic.

Metorial ties every agent action back to the agent and the person who delegated its access. You can always see who did what on whose behalf, with every grant logged and reversible.

Metorial supports secret encryption backed by AWS KMS, including your own KMS keys. You control the encryption of secrets stored in Metorial, with full audit logs.

Metorial automatically creates a Magic MCP server for each person. A virtual MCP endpoint they paste into their agent. They sign in with SSO once and reach every tool they're allowed to use, with no tokens, no secrets, and no config to manage.

Magic Network is the secure runtime your integrations run on. Every provider runs in its own isolated enclave with only the network access it needs, secrets stay in a managed vault backed by KMS, and you can run across multiple regions or on-prem.

Everything in Metorial can be done through the API, from integrations and access control to identities, skills, and Magic MCP, with SDKs for Node, Python, and Go.

Metorial Explorer lets you connect to any integration, explore its capabilities, and test it directly, with logs and performance in view so you understand how it works.

The Metorial Gateway routes every MCP connection in your organization through one governed path. Every tool call can be controlled, logged, and protected.

The Metorial Marketplace, delivered through , gives your organization one place to discover and use integrations and skills, with access controlled by your policies.

Metorial offers on-prem hosting for organizations with strict compliance needs, with full control over data and infrastructure, plus governance features for all customers.

Metorial Vault stores the credentials your integrations need securely and uses them without exposing them, with encryption keys managed by Metorial or by you and every operation audited.

Metorial monitors your integrations, traffic, and logs and alerts you when something needs attention. You can catch problems like schema changes and errors early.

Metorial offers EU and US regions so you can meet data residency requirements and give users fast performance, with on-prem hosting available for the strictest needs.

Metorial logs every MCP message, tool call, and execution in real time, giving you one place to monitor errors, performance, and usage across all your agents and integrations.

Metorial gives you granular, role-based access control over users, agents, and admins, with policies across skills, integrations, and agents, built on your existing identity.

Metorial Protoguard monitors every tool call, message, and execution for prompt injection attacks in real time. Attacks on your agents are detected before they do damage.

Metorial version-controls every integration and skill, tracking changes to tools, configs, and specs. You can see what changed, pin versions, or update automatically.

Providers are the reusable toolkits and integrations agents use in Metorial. Build or connect one once, govern it centrally, and make it available to the teams and agents allowed to use it.

Connect to MCP servers hosted by other vendors, like Linear, Sentry, Atlassian, and Microsoft, and bring them under the same access control, tracing, and monitoring as the rest of your providers.

SAML lets your people sign in once with your identity provider and access the MCP servers and skills their role allows, fully integrated with Metorial's access control.

Sandboxes are isolated, fully functional environments for development, staging, and testing. Try new integrations, skills, and configurations without affecting production, then promote what works.

Metorial version-controls every integration and skill and tracks changes to tools, configs, and schemas. An unexpected change is caught and alerted before it breaks an agent that depends on it.

Service accounts give scripts, pipelines, and backend systems their own governed identity in Metorial, scoped to the right project or environment and fully logged.

Session logs capture every tool call, message, and execution in an MCP connection. You can follow exactly what happened in a session and review it later.

Skills are reusable workflows that capture how a task should be done. Domain experts build them together in a collaborative document editor, share them across teams, and govern who can use them.

Skill export publishes the skills you build in Metorial as a Skill Marketplace on GitHub, GitLab, Bitbucket, or Azure DevOps. Others can install everything with a single git URL.

A skill marketplace is a centralized catalog of your team's skills, installable with a single URL and fully managed by Metorial, with access control and linked tools governed centrally.

Skill plugins package your Metorial skills so they work inside the agent tools your teams already use, like Claude Code, Codex, Cursor, and Antigravity, and stay in sync automatically.

SSO for providers lets people reach all the integrations and skills they're allowed to use with one login, without importing tokens or configuration for each one.

Metorial lets enterprises import existing credentials securely and export them when required, with both disabled by default, fully audited, and never readable by end users.

Tokenless auth lets agents and people reach integrations based on identity instead of API keys. There are no secrets to share, rotate, or leak.

Tools are the individual actions inside a provider. Metorial lets you grant or restrict them one at a time. An agent only gets the specific actions it needs.

Triggers let Metorial listen for events from your providers, like new messages or tickets, and start agent actions in response.

Metorial works with any model, agent, and provider. You can choose the best tools for each job and govern them all the same way, without locking into a single vendor.

Version pinning locks an integration to a version you trust. An upstream change can't break an agent without warning. Changes are still monitored and alerts keep you informed.