Back to vendors
M

Metorial

Also known as: Metorial Connect, Metorial Platform

Visit site
Entry priceFree Dev · Scale $250/mo incl. 20 seats + usage · Enterprise customFull pricing detail

Integration and governance platform connecting agents and MCP clients to 1,000+ tools, with per-user auth, triggers, portals, skills, Protoguard screening and full tool-call logs.

Metorial is an integration and governance platform that connects AI agents and MCP clients to the tools a company already uses. More than 1,000 catalog integrations are deployed as providers with per-user authentication, and an agent reaches their tools through a session's MCP URL, a Magic MCP URL that bundles approved tools behind company sign-in, or direct tool calls on the API. Customers add internal systems as custom providers deployed from code, container images or remote MCP servers, and provider triggers deliver events from external systems to the customer's application through metered callbacks.

Workforce is the control layer for teams. Portals publish approved integrations and Magic Skills, reusable workflows the customer's own experts write, to users by access group; people request more access and administrators resolve it; and Protoguard screens messages and tool calls for prompt injection and out-of-policy requests. Connection logs record every session and tool call with its arguments and result, tied to the identity behind it.

Metorial states SOC 2 Type 2, with reports under NDA, and GDPR compliance, offers SSO, SAML and SCIM, and runs in EU or US regions or on premises in the customer's cloud. Its core is source available under the Functional Source License (FSL-1.1-ALv2, which converts to Apache 2.0 later) and can be run on the customer's own infrastructure. It works with OpenAI, Anthropic, Google and OpenAI-compatible models through JavaScript and Python SDKs. Pricing starts with a free Dev plan, then Scale at $250 a month with 20 Workforce seats, and a quoted Enterprise plan.

Vendor details

Canonical URL

https://metorial.com

Category

Agent infrastructure

Funding status

Backed by Y Combinator; the site footer describes Metorial as part of Meteor Computing.

Company status

independent

Use cases & customers

Primary use cases

Connect agents and MCP clients to company toolsGovern agent tool access by user and groupPublish approved integrations and skills through portalsTrace agent tool calls

Target customers

Developers building agentic AI applicationsAI native companies shipping customer facing agentsEnterprises needing agent access governanceTeams hosting many MCP integrations

Deployment options

Cloud (hosted)Self hostedOn premises

Integrations

More than 1,000 catalog integrations with per-user OAuth, API-token or custom-credential auth; custom providers from code, container images or remote MCP servers; tools reached through session and Magic MCP URLs, the REST API or JavaScript and Python SDKs, formatted for OpenAI, Anthropic, Google, OpenAI-compatible models and the Vercel AI SDK; provider triggers with callbacks to the customer's application.

In practice

Your employees use Cursor, Claude Code and ChatGPT, and IT needs to control what those agents can reach. Metorial Workforce publishes approved integrations through a portal behind company SSO, with every tool call logged against the user.

Your product's agent needs to act in each customer's GitHub, Slack and HubSpot. Metorial deploys the integrations with per-user OAuth and returns an MCP URL for each session.

Your agent should react when something changes in a customer's tools. Provider triggers fire on external events, and Metorial's callbacks deliver them to your application.

Agentic Index coverage score

8.0 / 14 capabilities · 57%

Integrations & Tool Calling Full

More than 1,000 catalog integrations are available, each deployed as a provider with its own config and per user authentication (OAuth through Provider Setup Sessions, API tokens or custom credentials), and their tools reach the agent through a session's MCP URL or direct tool calls on the API. Customers add their own tools as custom providers deployed from code, container images or remote MCP servers, and can use their own OAuth apps.

Sourcemetorial.com/docs/build/operate/providersread 2026-09-22

Workflow Orchestration Not documented

Agents connect to tools through Metorial, which does not run the agent itself: sessions merge the tools of several providers behind one connection, and Magic Skills package instructions, files and optional scripts that the customer's AI client or agent follows when it invokes the skill. Sequencing and branching happen in that client or in the customer's framework, and no workflow runtime of Metorial's own is documented.

Sourcemetorial.com/docs/platform/workforce/magic-skillsread 2026-09-22

Knowledge Grounding & RAG Not documented

Live access to the systems that hold company knowledge (Google Drive, Notion, Confluence and others through its integrations) is what Metorial gives agents, and a Magic Skill can carry reference documents for its instructions. No index, retrieval layer or semantic search over the customer's corpus is documented, and the API's Documents and Stores resources are not described as retrieval.

Sourcemetorial.com/docs/platform/get-started/core-conceptsread 2026-09-22

Human Oversight & Guardrails Partial

Several controls constrain what agents may do: access groups decide which integrations and skills each user or SSO group sees, people request more access and administrators review and resolve those requests, session tool filters can restrict a provider to read only tools, and Protoguard screens each message and tool call, blocking prompt injection and requests outside the customer's policies. These are constraints on the agent and approvals of access; no approval step a person acts on before a specific agent action commits is documented.

Sourcemetorial.com/docs/platform/workforce/portalsread 2026-09-22

Security, Identity & Governance Full

Metorial states SOC 2 Type 2, independently audited for security, availability and confidentiality with reports under NDA, and GDPR compliance with a DPA and a published subprocessor list, and the homepage carries a SOC 2 badge. The access side is specific: SSO and SAML with per user and per group policies, roles as named permission sets attached to members and teams, SCIM and SAML IdP sign in on the homepage, and secrets encrypted in Vault with keys managed through KMS; RBAC and SAML SSO are listed on the Enterprise plan.

Sourcemetorial.com/securityread 2026-09-22

Observability & Auditability Full

Connection logs record what agents did: each session's timeline, and every tool call opened to its tool name, arguments and result, filtered across sessions by status, source and agent, with tool errors listed separately. Workforce shows each agent's operations and connection history tied to its linked identity, the logs also carry provider auth and Protoguard events, and the Tool Calls API exposes inputs, outputs and status. Log retention is 30 days on Dev, 90 on Scale and custom on Enterprise.

Sourcemetorial.com/docs/platform/integrations/review-connection-logsread 2026-09-22

Memory & State Persistence Not documented

Sessions are lightweight runtime contexts created per request or per user interaction, and Metorial stores each user's provider credentials as auth configs for reuse, but no memory layer that keeps an agent's context across runs is documented. Stored credentials and a session's connection record are not memory with a stated scope and lifetime.

Sourcemetorial.com/docs/build/operate/sessionsread 2026-09-22

Deployment & Data Residency Full

Deployment can sit in Metorial's cloud or the customer's: the Security page offers on premises deployment in the customer's own cloud or data center and a choice of EU or US regions for data residency, and each provider runs in its own isolated enclave behind firewalls. On premises deployment is listed on the Enterprise plan.

Sourcemetorial.com/securityread 2026-09-22

Prebuilt Agents / Templates / Packs Not documented

The machinery to package a customer's own workflows is what Metorial supplies: Magic Skills with templates, groups and curated skill marketplaces, imports from public repositories, and portals that publish them to users. The skills in those catalogs are written by the customer's own experts, the integration catalog is a set of integrations rather than packaged work, and the three sample apps in the docs are tutorial code. No ready made skill, template or role specific agent of Metorial's own is documented.

Sourcemetorial.com/docs/platform/workforce/magic-skillsread 2026-09-22

Triggers & Channel Coverage Full

Providers expose triggers that fire on events in the external system, and a callback, a webhook Metorial registers against every matching integration instance, notifies the customer's application so it can run event driven workflows (glossary and Callbacks API reference). Callbacks are a metered plan feature, 10,000 a month on Dev and 100,000 on Scale. Events in outside systems can start work without a person asking.

Sourcemetorial.com/docs/build/glossaryread 2026-09-22

Model Flexibility & Routing Full

Tools reach whatever model the customer runs: the provider runtime formats session tools for OpenAI, Anthropic, Google Gemini and OpenAI compatible APIs such as DeepSeek, Together AI and xAI, and for the Vercel AI SDK, each called with the customer's own provider key, and any MCP client such as Cursor, Claude Code or Codex can connect. Metorial names no model of its own, so model choice, provider and key are the customer's throughout.

Sourcemetorial.com/docs/build/providersread 2026-09-22

APIs / SDKs / MCP Extensibility Full

A documented REST API with keys scoped to one environment makes Metorial callable from outside, covering providers, deployments, sessions, tool calls, skills, portals and callbacks, alongside JavaScript and Python SDKs and a CLI. It serves MCP as addressed endpoints, a session connection URL and Magic MCP URLs, and customers package their own tools as versioned custom providers with rollback.

Sourcemetorial.com/docs/build/apiread 2026-09-22

Testing, Debugging & Optimization Partial

A testing path exists before production: sandbox environments with their own deployments, keys, auth configs and logs, where a development key never reaches production, Metorial Explorer for calling a provider's tools against test data, and version pinning and instant rollback for custom providers. That is validation and debugging before release; no way to evaluate the customer's agent with a readable, comparable result (scored test cases, a pass rate or a judge verdict) is documented.

Sourcemetorial.com/docs/build/operate/organizations-and-projectsread 2026-09-22

Browser / Computer-use Not documented

Everything runs through APIs and MCP servers: Metorial's providers reach applications that way, and no browser, desktop or remote machine control of its own is documented.

Sourcemetorial.com/docs/build/operate/providersread 2026-09-22

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Pricing

Free Dev · Scale $250/mo incl. 20 seats + usage · Enterprise custom

Monthly plan fee with included tool calls, callbacks and Workforce seats, plus per-seat and usage overage

Free tierTrial available

Included quota

Dev: 500,000 tool calls, 10,000 callbacks, 2 Workforce seats, 10 provider integrations, 1 project and environment, 30-day log retention. Scale: 2.5 million tool calls, 100,000 callbacks, 20 Workforce seats, unlimited integrations, projects and environments, 90-day log retention.

What is public

Every self-serve tier is priced with its allowances and overage rates; only Enterprise is quoted.

Billing mechanics

Sign up without a sales call on the free Dev plan, move to Scale at a monthly fee that includes seats and usage allowances with per-seat and per-call overage, or buy Enterprise on a quote.

Cost watchouts

Seats beyond the 20 included on Scale cost $20 each, and tool calls and callbacks past the plan allowance bill at $1 per 1,500 and $1 per 750. RBAC, SAML SSO and on-premises deployment sit on the quoted Enterprise plan, and unlimited items are subject to fair use.

Variable cost rationale

Tool calls, callbacks and seats each have an allowance and a published overage rate, so cost rises with agent activity and headcount past the Scale allowances; the free Dev allowance of 500,000 tool calls covers light use.

Overage / add-ons

$1 per 1,500 additional tool calls and $1 per 750 additional callbacks on Scale; custom on Enterprise. Unlimited items are subject to fair use.

Sales call required

Mixed (some tiers require a call)

Free / trial

Metorial Dev free: 500K tool calls, 10K callbacks, 2 seats, 10 integrations; free Workforce trial

Lowest paid plan

Metorial Scale, $250 a month including 20 Workforce seats, 2.5 million tool calls and 100,000 callbacks, $20 per additional seat

Commercial notes

Backed by Y Combinator and part of Meteor Computing. Enterprise adds on-premises deployment, SOC 2 and GDPR compliance documentation, RBAC, SAML SSO and Slack or Teams support with an SLA.

Key ambiguities

Unlimited sessions, connections and auth connections are subject to fair use, and Metorial may ask a heavy user to upgrade or limit usage. Enterprise terms are quoted. Third-party directories have cited HIPAA, which is not stated on Metorial's own pages.

Agentic Index verified 2026-09-22

Alternatives to Metorial

The closest documented capability profiles to Metorial among agent infrastructure platforms tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

  • Freeplay8.0 / 14Fuller documented coverage on Testing, Debugging & Optimization
  • Arcade7.5 / 14Fuller documented coverage on Testing, Debugging & Optimization
  • Composio8.5 / 14Adds documented Workflow Orchestration and Memory & State PersistenceMetorial vs Composio →
  • Confident AI8.5 / 14Adds documented Prebuilt Agents, Templates & Packs
  • Fiddler AI8.5 / 14Fuller documented coverage on Human Oversight & Guardrails and Testing, Debugging & Optimization
  • Helicone7.5 / 14Fuller documented coverage on Human Oversight & Guardrails

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.