Barndoor
Barndoor AI is the control plane for agentic AI, providing secure access and governance for AI agents and Model Context Protocol (MCP) servers. Founded in 2024 by Oren Michels (founder of Mashery), Barndoor enables enterprise IT, security, and developer teams to register agents, govern MCP server access through policy, broker OAuth connections to backend SaaS, and proxy MCP traffic with runtime policy enforcement and full audit trails. The Barndoor Platform REST API manages servers, connections, policies, agents, and MCP / SSE request proxying. Python, TypeScript, and Go SDKs are published on GitHub alongside Rust SDKs (Cerbos, official MCP, MCP OAuth compliance suite) and a Crew AI example. Deployment options include SaaS (trial), private cloud, and on-premises (Enterprise).
8 APIs
14 Features
AI AgentsAI GovernanceAgentic AIMCPModel Context ProtocolPolicy EnforcementOAuthIdentitySecurityAuditControl Plane
REST API for the Barndoor Platform. Manage MCP server registrations, OAuth connections from agents to backend SaaS, access-control policies (with rules, restrictions, revisions,...
Python SDK for the Barndoor AI Platform. Wraps the Platform REST API, handles Auth0 PKCE login (`loginInteractive()`), discovers governed MCP tools, brokers OAuth connections to...
TypeScript SDK for the Barndoor AI Platform. Browser- and Node-friendly client for Auth0 PKCE login, governed MCP tool discovery, OAuth connection initiation, and proxying MCP /...
Go SDK for the Barndoor AI Platform. Server-side client for registering agents, managing MCP servers and policies, brokering OAuth connections, and proxying MCP requests from Go...
The official Rust SDK for the Model Context Protocol. Maintained under the Barndoor AI GitHub organization; provides primitives to build MCP clients and servers in Rust.
Rust SDK for Cerbos, the policy-decision-point used by Barndoor for attribute-based access control. Lets Rust services request policy decisions from a Cerbos PDP.
Rust test suite that validates remote MCP servers against the MCP authorization specification - RFC 9728 (Protected Resource Metadata), RFC 8414 (Authorization Server Metadata),...
Reference Python demo application showing how to plug Barndoor-governed MCP tools into a Crew AI multi-agent workflow.
MCP Governance
Secure access control and policy enforcement for Model Context Protocol servers.
Runtime Policy Enforcement
Continuous governance applied at the moment AI agents act, not just at login.
Right-Sized Permissions
Precise, scoped access for agents - not broad human-level permissions.
Context Filtering
Dynamically surface only policy-compliant MCP tools, optimizing the context window.
AI Agent Registry
Register internal and external agents, group them, and track activity.
OAuth Connection Brokering
Initiate and manage OAuth 2.0 connections from agents to backend SaaS.
MCP / SSE Proxying
Streaming proxy that injects credentials and enforces policy on every MCP and SSE request.
Policy Authoring (RBAC/ABAC)
Create, clone, version, validate, and apply Cerbos-based RBAC and ABAC policies.
Audit Dashboards and Activity Logs
Complete audit trails for every AI action, applied policy, and outcome.
Audit Log Export
Stream audit events as gzipped JSON Lines to S3 / GCS / MinIO / SeaweedFS buckets.
Shadow AI Discovery
Centralized visibility into unauthorized AI apps and agents in the environment.
Identity Provider Integration
Connect to existing enterprise IdPs (Keycloak-based) for SSO and identity.
Static Egress IPs
Five dedicated outbound IPs for whitelisting Barndoor traffic at MCP servers.
Private and On-Prem Deployment
SaaS, private cloud, and on-premises deployment options for sensitive environments.
Enterprise AI Governance
Apply access policies and governance to AI agents across the organization.
MCP Server Management
Centrally register, secure, and manage MCP server deployments for AI agents.
Agentic Workflow Orchestration
Coordinate multi-agent workflows with security and accountability controls.
AI Security and Data Exfiltration Prevention
Prevent unauthorized AI agent actions and limit data exfiltration.
Shadow AI Discovery
Surface unauthorized AI apps and agents already running in the environment.
Developer Tooling for Governed Agents
Build agents safely with end-to-end policy enforcement via SDKs.
Microsoft 365 Agent Governance
Govern agents that work across Microsoft 365 (Excel, Outlook, Teams, OneDrive).
IT & Security Teams
Centralize AI governance, manage shadow AI, and enforce real-time access controls at scale.
Developers
Deploy agents safely without custom security logic, with end-to-end policy across dev, staging, and prod.
aid: barndoor
url: https://raw.githubusercontent.com/api-evangelist/barndoor/refs/heads/main/apis.yml
name: Barndoor
kind: company
description: >-
Barndoor AI is the control plane for agentic AI, providing secure access and governance for AI agents and Model
Context Protocol (MCP) servers. Founded in 2024 by Oren Michels (founder of Mashery), Barndoor enables enterprise IT,
security, and developer teams to register agents, govern MCP server access through policy, broker OAuth connections to
backend SaaS, and proxy MCP traffic with runtime policy enforcement and full audit trails. The Barndoor Platform REST
API manages servers, connections, policies, agents, and MCP / SSE request proxying. Python, TypeScript, and Go SDKs
are published on GitHub alongside Rust SDKs (Cerbos, official MCP, MCP OAuth compliance suite) and a Crew AI example.
Deployment options include SaaS (trial), private cloud, and on-premises (Enterprise).
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
tags:
- AI Agents
- AI Governance
- Agentic AI
- MCP
- Model Context Protocol
- Policy Enforcement
- OAuth
- Identity
- Security
- Audit
- Control Plane
created: '2026-03-16'
modified: '2026-05-19'
specificationVersion: '0.19'
apis:
- aid: barndoor:platform-api
name: Barndoor Platform API
description: >-
REST API for the Barndoor Platform. Manage MCP server registrations, OAuth connections from agents to backend
SaaS, access-control policies (with rules, restrictions, revisions, validation), AI agent registrations, and proxy
live MCP requests (`/mcp/{server_name}`) and SSE streams (`/sse/{server_name}`) through Barndoor's policy
enforcement and audit pipeline. JWT Bearer authentication via Auth0 OAuth 2.0 with PKCE; the SDK's
`loginInteractive()` handles the OAuth flow.
humanURL: https://docs.barndoor.ai/api-reference/introduction
baseURL: https://{organization_id}.platform.barndoor.ai
tags:
- Platform API
- MCP
- Policy
- Agents
- Servers
- Connections
- Proxy
properties:
- type: Documentation
url: https://docs.barndoor.ai/api-reference/introduction
- type: OpenAPI
url: openapi/barndoor-openapi.yml
- type: Authentication
url: https://docs.barndoor.ai/api-reference/introduction
- type: SDK
url: https://docs.barndoor.ai/sdks/introduction
- aid: barndoor:python-sdk
name: Barndoor Python SDK
description: >-
Python SDK for the Barndoor AI Platform. Wraps the Platform REST API, handles Auth0 PKCE login
(`loginInteractive()`), discovers governed MCP tools, brokers OAuth connections to backend SaaS, and exposes the
catalog through Pythonic helpers compatible with OpenAI tool-calling and frameworks such as Crew AI.
humanURL: https://docs.barndoor.ai/sdks/python
baseURL: https://github.com/barndoor-ai/barndoor-python-sdk
tags:
- Python SDK
- SDK
- MCP
properties:
- type: Documentation
url: https://docs.barndoor.ai/sdks/python
- type: Repository
url: https://github.com/barndoor-ai/barndoor-python-sdk
- aid: barndoor:typescript-sdk
name: Barndoor TypeScript SDK
description: >-
TypeScript SDK for the Barndoor AI Platform. Browser- and Node-friendly client for Auth0 PKCE login, governed MCP
tool discovery, OAuth connection initiation, and proxying MCP / SSE requests through Barndoor.
humanURL: https://docs.barndoor.ai/sdks/typescript
baseURL: https://github.com/barndoor-ai/barndoor-ts-sdk
tags:
- TypeScript SDK
- SDK
- MCP
properties:
- type: Documentation
url: https://docs.barndoor.ai/sdks/typescript
- type: Repository
url: https://github.com/barndoor-ai/barndoor-ts-sdk
- aid: barndoor:go-sdk
name: Barndoor Go SDK
description: >-
Go SDK for the Barndoor AI Platform. Server-side client for registering agents, managing MCP servers and policies,
brokering OAuth connections, and proxying MCP requests from Go services.
humanURL: https://github.com/barndoor-ai/barndoor-go-sdk
baseURL: https://github.com/barndoor-ai/barndoor-go-sdk
tags:
- Go SDK
- SDK
- MCP
properties:
- type: Repository
url: https://github.com/barndoor-ai/barndoor-go-sdk
- aid: barndoor:official-mcp-rust-sdk
name: Official MCP Rust SDK
description: >-
The official Rust SDK for the Model Context Protocol. Maintained under the Barndoor AI GitHub organization;
provides primitives to build MCP clients and servers in Rust.
humanURL: https://github.com/barndoor-ai/official-mcp-rust-sdk
baseURL: https://github.com/barndoor-ai/official-mcp-rust-sdk
tags:
- MCP
- Rust
- SDK
properties:
- type: Repository
url: https://github.com/barndoor-ai/official-mcp-rust-sdk
- aid: barndoor:cerbos-sdk-rust
name: Cerbos Rust SDK
description: >-
Rust SDK for Cerbos, the policy-decision-point used by Barndoor for attribute-based access control. Lets Rust
services request policy decisions from a Cerbos PDP.
humanURL: https://github.com/barndoor-ai/cerbos-sdk-rust
baseURL: https://github.com/barndoor-ai/cerbos-sdk-rust
tags:
- Cerbos
- ABAC
- Policy
- Rust
- SDK
properties:
- type: Repository
url: https://github.com/barndoor-ai/cerbos-sdk-rust
- aid: barndoor:mcp-auth-compliance
name: MCP OAuth Compliance Suite
description: >-
Rust test suite that validates remote MCP servers against the MCP authorization specification - RFC 9728
(Protected Resource Metadata), RFC 8414 (Authorization Server Metadata), RFC 7591 (Dynamic Client Registration),
and OAuth 2.1. Useful for vendors and customers verifying MCP server conformance before onboarding to Barndoor.
humanURL: https://github.com/barndoor-ai/mcp-auth-compliance
baseURL: https://github.com/barndoor-ai/mcp-auth-compliance
tags:
- MCP
- OAuth
- Compliance
- Rust
- Conformance
properties:
- type: Repository
url: https://github.com/barndoor-ai/mcp-auth-compliance
- aid: barndoor:crew-ai-example
name: Barndoor + Crew AI Example
description: >-
Reference Python demo application showing how to plug Barndoor-governed MCP tools into a Crew AI multi-agent
workflow.
humanURL: https://github.com/barndoor-ai/barndoor-ai-crew-ai-python-example
baseURL: https://github.com/barndoor-ai/barndoor-ai-crew-ai-python-example
tags:
- Crew AI
- Python
- Example
- MCP
properties:
- type: Repository
url: https://github.com/barndoor-ai/barndoor-ai-crew-ai-python-example
common:
- type: LinkedIn
url: https://www.linkedin.com/company/barndoor-ai
- type: Website
url: https://barndoor.ai/
name: Barndoor AI
- type: Documentation
url: https://docs.barndoor.ai/
name: Barndoor Developer Documentation
- type: APIReference
url: https://docs.barndoor.ai/api-reference/introduction
name: Barndoor API Reference
- type: OpenAPI
url: openapi/barndoor-openapi.yml
name: Barndoor Platform API OpenAPI
- type: Authentication
url: https://docs.barndoor.ai/api-reference/introduction
name: Authentication (Auth0 OAuth 2.0 with PKCE)
- type: SDK
url: https://docs.barndoor.ai/sdks/introduction
name: Barndoor SDKs (Python, TypeScript, Go)
- type: Portal
url: https://app.barndoor.ai/
name: Barndoor App
- type: SignUp
url: https://app.barndoor.ai/auth/signup/trial
name: Barndoor Free Trial Signup
- type: TokensManagement
url: https://app.barndoor.ai/settings/tokens
name: Platform API Tokens
- type: Pricing
url: https://barndoor.ai/pricing
name: Barndoor Pricing
- type: Plans
url: plans/barndoor-plans-pricing.yml
name: Barndoor Plans (API Commons)
- type: RateLimits
url: rate-limits/barndoor-rate-limits.yml
name: Barndoor Rate Limits (API Commons)
- type: FinOps
url: finops/barndoor-finops.yml
name: Barndoor FinOps (FOCUS 1.3)
- type: GitHub
url: https://github.com/barndoor-ai
name: Barndoor AI GitHub Org
- type: Security
url: https://barndoor.ai/security/
name: Barndoor Security
- type: TrustCenter
url: https://trust.barndoor.ai
name: Barndoor Trust Center
- type: About
url: https://barndoor.ai/about-us/
name: About Barndoor AI
- type: MCPCatalog
url: https://docs.barndoor.ai/mcp-servers/servers
name: Barndoor MCP Catalog (60+ servers)
- type: IPAllowlist
url: https://docs.barndoor.ai/how-tos/ip-whitelisting
name: Static Egress IPs for MCP Servers
- type: LogExport
url: https://docs.barndoor.ai/how-tos/log-export
name: Audit Log Export to S3-Compatible Storage
- type: SpectralRules
url: rules/barndoor-spectral-rules.yml
name: Spectral Ruleset
- type: Vocabulary
url: vocabulary/barndoor-vocabulary.yaml
name: Barndoor Vocabulary
- type: JSONLD
url: json-ld/barndoor-context.jsonld
name: Barndoor JSON-LD Context
- name: Features
type: Features
data:
- name: MCP Governance
description: Secure access control and policy enforcement for Model Context Protocol servers.
- name: Runtime Policy Enforcement
description: Continuous governance applied at the moment AI agents act, not just at login.
- name: Right-Sized Permissions
description: Precise, scoped access for agents - not broad human-level permissions.
- name: Context Filtering
description: Dynamically surface only policy-compliant MCP tools, optimizing the context window.
- name: AI Agent Registry
description: Register internal and external agents, group them, and track activity.
- name: OAuth Connection Brokering
description: Initiate and manage OAuth 2.0 connections from agents to backend SaaS.
- name: MCP / SSE Proxying
description: Streaming proxy that injects credentials and enforces policy on every MCP and SSE request.
- name: Policy Authoring (RBAC/ABAC)
description: Create, clone, version, validate, and apply Cerbos-based RBAC and ABAC policies.
- name: Audit Dashboards and Activity Logs
description: Complete audit trails for every AI action, applied policy, and outcome.
- name: Audit Log Export
description: Stream audit events as gzipped JSON Lines to S3 / GCS / MinIO / SeaweedFS buckets.
- name: Shadow AI Discovery
description: Centralized visibility into unauthorized AI apps and agents in the environment.
- name: Identity Provider Integration
description: Connect to existing enterprise IdPs (Keycloak-based) for SSO and identity.
- name: Static Egress IPs
description: Five dedicated outbound IPs for whitelisting Barndoor traffic at MCP servers.
- name: Private and On-Prem Deployment
description: SaaS, private cloud, and on-premises deployment options for sensitive environments.
- name: Use Cases
type: UseCases
data:
- name: Enterprise AI Governance
description: Apply access policies and governance to AI agents across the organization.
- name: MCP Server Management
description: Centrally register, secure, and manage MCP server deployments for AI agents.
- name: Agentic Workflow Orchestration
description: Coordinate multi-agent workflows with security and accountability controls.
- name: AI Security and Data Exfiltration Prevention
description: Prevent unauthorized AI agent actions and limit data exfiltration.
- name: Shadow AI Discovery
description: Surface unauthorized AI apps and agents already running in the environment.
- name: Developer Tooling for Governed Agents
description: Build agents safely with end-to-end policy enforcement via SDKs.
- name: Microsoft 365 Agent Governance
description: Govern agents that work across Microsoft 365 (Excel, Outlook, Teams, OneDrive).
- name: Solutions
type: Solutions
data:
- name: IT & Security Teams
description: Centralize AI governance, manage shadow AI, and enforce real-time access controls at scale.
- name: Developers
description: Deploy agents safely without custom security logic, with end-to-end policy across dev, staging, and prod.
- name: Compliance
type: Compliance
data:
- name: SOC 2 Type II
description: Barndoor holds a SOC 2 Type II attestation for security controls effectiveness over time.
- type: LLMsTxt
url: https://docs.barndoor.ai/llms.txt
integrations:
- name: Salesforce
- name: Notion
- name: GitHub
- name: GitLab
- name: Slack
- name: HubSpot
- name: Microsoft 365
- name: Microsoft Teams
- name: Microsoft Excel
- name: Microsoft Word
- name: OneDrive
- name: OneNote
- name: PowerPoint
- name: Outlook Mail
- name: Outlook Calendar
- name: Microsoft Planner
- name: Microsoft Dynamics
- name: SharePoint
- name: Gmail
- name: Google Calendar
- name: Google Docs
- name: Google Sheets
- name: Google Slides
- name: Google Drive
- name: Atlassian
- name: Linear
- name: Asana
- name: Monday
- name: Basecamp
- name: Aha!
- name: Box
- name: Dropbox
- name: Figma
- name: Airtable
- name: Snowflake
- name: Hex
- name: Amplitude
- name: SonarQube
- name: Datadog
- name: Grafana
- name: Sentry
- name: Harness
- name: Finch
- name: ServiceNow
- name: Zendesk
- name: Freshdesk
- name: Intercom
- name: Zoom
- name: Fireflies.ai
- name: Granola
- name: Otter
- name: Apollo
- name: Attio
- name: Close
- name: Gong
- name: Shopify
- name: Zoho CRM
- name: Stripe
- name: Plaid
- name: QuickBooks
- name: Xero
- name: Octagon
- name: Crew AI
- name: Auth0
- name: Keycloak
- name: Cerbos
maintainers:
- FN: Kin Lane
email: kin@apievangelist.com