Macha
Macha Concepts

Custom Tools

Definition

In Macha, a custom tool lets an AI agent call any HTTP API endpoint you configure — so an agent can read live data or take an action in a system Macha doesn't have a built-in connector for, without waiting for a marketplace app.

Also known as: custom actionscustom API toolsagent tools

How it works

You define the tool once as an org-level resource: the request method (GET/POST/PUT/PATCH/DELETE), the endpoint URL with {{param}} placeholders, authentication (API key, bearer token, or basic auth — credentials are encrypted and never exposed to the model), the parameters the AI should supply, an optional JSON body template, and a response mapping that trims the payload down to what the agent needs.

Each tool is typed Read or Write. Read tools just fetch data; write tools create, update, or delete — and, like connector write tools, pause for confirmation in interactive chat before they run. You then assign the tool to whichever agents need it.

Why it matters

Custom tools are what let an agent act on your stack — an internal CRM, a logistics provider, a niche SaaS — not just the systems Macha ships connectors for. They are the practical expression of agentic AI for the tools your team already uses.

You can build them by hand or describe the API to Sidekick, which discovers endpoints, tests your credentials, and wires the tool up for you.

Frequently asked

Do I need a built-in connector for every tool?

No — that's the point. If a system exposes an HTTP API, you can point a Macha agent at it directly with a custom tool instead of waiting for a native connector.

Are custom tools available on every plan?

Custom tools are a Professional and Enterprise feature, metered by assignment slots — assigning the same tool to 3 agents counts as 3 slots (5 slots on Professional, 20 on Enterprise).

Put these ideas to work

Macha is an AI agent layer that sits on top of the help desk you already run — Zendesk, Freshdesk, Front, Intercom, or Gorgias.

Start Trial