Zendesk
Freshdesk
Gorgias
Front
Shopify
Stripe
Slack
Notion
Google Workspace
Confluence
Plain-English Agent Configuration
Definition
In Macha, you configure an agent by writing its rules in plain English in its instructions field — how to triage, what tone to use, what to escalate, what's off-limits — rather than coding logic. The model obeys those instructions on every conversation.
How it works
An agent's instructions are its system prompt — the single most important part of its configuration. You describe the agent's role, persona, boundaries, and workflow in ordinary language: "You are a Tier 1 support agent for Acme. Read the ticket, look up the customer's order, draft a reply. Never promise a refund without checking billing." Alongside the instructions you pick a model and enable the specific tools and data sources the agent may use.
You don't have to trust that the model reads your rules perfectly, because the platform enforces the hard limits: tools you didn't enable are invisible to the agent, and write tools pause for confirmation in chat. The instructions shape behavior; the platform enforces boundaries.
Why it matters
Plain-English configuration is what makes agents accessible to support and ops teams, not just engineers. And it's measurable: Agent Evaluation runs an AI judge over past conversations to score how well the agent actually followed those written instructions, so you can tell whether a prompt change helped.
Frequently asked
Do I have to write instructions from scratch?
No. You can start from a template or describe the agent to the builder / Sidekick, which drafts the instructions for you — then edit the plain-English text to fit your process.
If it's just plain English, what stops the agent going rogue?
The platform enforces limits regardless of the prompt: disabled tools can't be called, and write actions pause for confirmation in interactive chat. Instructions guide behavior; the platform guards it.
Related terms
AI Agent Builder
The AI Agent Builder is Macha's Build-with-AI flow: describe the agent you want in plain language and it configures the name, instructions, tools, triggers, and even sub-agents based on the integrations you've connected..
System Prompt
A system prompt is the underlying set of instructions that defines an AI model's role, behavior, tone, and boundaries for a conversation — set by the developer, not the end user, and applied to every interaction..
Prompt Engineering
Prompt engineering is the practice of designing and refining the instructions given to a language model to get accurate, relevant, and reliable outputs for a specific task..
Agent Templates
Agent templates are pre-built Macha agent configurations — name, instructions, recommended connectors, and pre-selected tools — that spin up a working agent for a common workflow in one click..
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