Squadico
Get started

Feature

AI agents that act in your tools, not just talk

Connect your business apps through MCP servers, your own APIs or ready-made tools. Your agents look things up, create and update, and you decide which actions need a person's approval.

Create my first agent

What connecting tools changes

  • MCP servers in a few clicks

    Declare any server speaking the Model Context Protocol (MCP), authorize it, and its tools become available to your agents.

  • Your own APIs as tools

    Describe a call to your API, or import an OpenAPI description, and each operation becomes a tool an agent can use.

  • Permissions per action

    For every tool, choose whether it runs on its own, waits for a person's approval or is blocked. An agent can only narrow what your organization allows.

  • Acting as the right person

    A connector can use your organization's account, each member's own account, or the identity of the user signed in on your site.

Give an agent its tools in three steps

  1. Declare the tool

    Add an MCP server, import an API description or turn on a built-in tool for your organization.

  2. Set its permission

    Choose whether each tool is allowed, requires approval or is blocked.

  3. Bind it to an agent

    Give the agent the tools it needs. It calls them when a request requires it, within the limits you set.

What agents do with tools

  • Look up a record

    Find an order, a ticket or a customer in your apps and answer with the real data.

  • Create and update

    Open a ticket, update a deal or schedule a meeting, with approval for the actions that matter.

  • Send an email

    Draft and send a message from your organization's mailbox, after a person has approved it.

Frequently asked questions

What is MCP?

The Model Context Protocol, an open standard for exposing tools to AI agents. A server that speaks MCP can be connected to Squadico without custom code.

Can I connect my own API?

Yes. Describe the call yourself, or import an OpenAPI description and choose which operations become tools.

How do I stop an agent from doing something risky?

Mark the tool as requiring approval or block it. Your organization sets the ceiling, and an agent can only be stricter.

Whose account does the agent use?

You choose per connector: the organization's account, each member's own account, or the account of the user signed in on your site.

Keep exploring

Ready to put your first agent to work?

Create your account, describe what your agent should do and publish it on your first channel today.

Create my first agent