Permify

Permify MCP Connector for Claude

F

Manage fine-grained authorization and access control via Permify — write schemas, manage relation tuples, and perform real-time permission checks.

21 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect your Permify instance to any AI agent to orchestrate complex authorization models and access control policies through natural language.

What you can do

  • Schema Management — Write, read, and partially update authorization models (DSL) for specific tenants using write_schema and partial_write_schema.
  • Data Handling — Manage relation tuples and attributes via write_data to define how users relate to resources.
  • Permission Checks — Perform resource-based authorization checks with check_permission, or execute bulk_check_permissions for high-scale validation.
  • Relationship Auditing — Query stored relational tuples with read_relationships and expand permissions to see who has access to what.
  • Multi-Tenancy — Create, list, and delete isolated tenants to manage multiple client environments from a single interface.

How it works

  1. Subscribe to this server
  2. Enter your Permify API URL and Secret Token
  3. Start managing your IAM infrastructure from Claude, Cursor, or any MCP-compatible client

Who is this for?

  • Security Engineers — Audit and verify access control logic and relationship graphs without manual API calls
  • Backend Developers — Quickly update schemas and test relationship tuples during development cycles
  • DevOps Teams — Automate tenant provisioning and policy bundle deployments through the agent
iamauthorizationrbacabacrebacaccess-control

21 tools expose this connector's capabilities to your AI agent.

bulk_check_permissions

Check multiple permissions in a single request

check_permission

Perform a resource-based authorization check

create_tenant

Create a new Permify tenant

delete_bundle

Delete a predefined bundle

delete_data

Delete relation tuples and attributes

delete_tenant

Delete a Permify tenant

expand_permissions

Retrieve all subjects that have a relationship with an entity

list_schemas

List all schema versions for a tenant

list_tenants

List all Permify tenants

lookup_entity

Find which resources a user can perform an action on

lookup_subject

Find which subjects can do an action on an entity

partial_write_schema

Partially update an authorization model

read_attributes

Query stored attributes

read_bundle

Retrieve details of a specific bundle

read_relationships

Query stored relational tuples

read_schema

Inspect a specific schema version

run_bundle

Execute a predefined bundle

subject_permission

List which permissions a user can perform on an entity

write_bundle

Define or update data bundles

write_data

Create relation tuples and attribute data

write_schema

Write an authorization model schema

See how to talk to your AI agent using Permify.

Check if user 'user:123' has 'edit' permission on 'document:doc-001' in tenant 'default'.

I've performed the check using `check_permission`. The result is 'ALLOWED'. User 123 has the necessary edit rights for that document.

List all schema versions for tenant 'production-v1'.

Fetching schema history... I found 3 versions for 'production-v1'. The latest version is 'v20231027-01'. Would you like to inspect its details using `read_schema`?

Create a new tenant called 'client-alpha-test'.

Executing `create_tenant`... Success! The tenant 'client-alpha-test' has been created with ID 'cat-9920'. You can now start writing its authorization schema.

Yes! Use the `bulk_check_permissions` tool. It allows you to send a list of check requests in a single call, making it efficient for auditing complex views or batch processing.

Related Connectors