Extensibility - MCP

Your AI client, your access.

  • One personal token, shown once and stored only as a hash.
  • Tools run under your current access, never a new role.
  • Expire or revoke a connection whenever you need to.

Illustrative product view. Sample data is fictional and permission labels are educational.

How it works

Concrete product states, permissions, and handoffs

Each demonstration uses fictional sample data and keeps the source record, actor, permission boundary, and next step visible.

One personal connection, not a shared super-token

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. One personal connection, not a shared super-token is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

The token does not become a new role

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. The token does not become a new role is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Tools expose actions, not direct database access

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. Tools expose actions, not direct database access is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Create, copy once, expire, revoke

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. Create, copy once, expire, revoke is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Live access changes apply to future calls

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. Live access changes apply to future calls is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Rate limits and audit context

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. Rate limits and audit context is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Choose the client and configure the endpoint

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. Choose the client and configure the endpoint is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

When MCP is not the right integration

This section explains the concrete operating path in Perelan: the actor, the record, the state change, and the permission boundary stay visible instead of becoming a loose note in chat. When MCP is not the right integration is described through a fictional but product-grounded state so buyers can understand what changes in the workflow and where the limits are.

Questions

What buyers usually ask

Answers are scoped to repository-verified behavior or intentionally point to current customer documentation when policy facts are not public.

Model Context Protocol lets an external AI client call product tools through a defined server interface.

Next step

Build a people operation your team will actually use.

Start with the workflow you want to make calmer, then map the people, records, permissions, and handoffs that make it real.