AmirahCo AI guide.

What CLI does.

AmirahCo AI is home to our AI tools. AmirahCo CLI is the app that runs a team of AI agents to build software.

An AI agent can do jobs such as writing code or testing a page. CLI plans their jobs and keeps the team working on the same project.

Agents can build different parts at once. The goal is to move your project forward faster, without making you manage each agent.

You set the goal. CLI runs the team.

You describe what you want to make and answer the app’s questions. You try the result and ask for changes.

CLI makes the plan. CLI gives each agent a job and keeps their work in sync. It brings the parts together and checks them.

From an idea to a result.

1. Describe the project. Tell the app what you need. For example, a business website with a contact form.
2. Let CLI run the team. The app splits the work. One agent can build the pages while another builds the form.
3. Try what they build. Use the site and tell the app what needs to change. CLI sends that work back to the team.

Some jobs need to wait for others. CLI tracks that order so the parts fit into one project.

Downloads and setup.

AmirahCo CLI is in development. Downloads and setup instructions will be shared when the app is ready to try.

Building your own AI tool?

An API lets two programs talk to each other. An AI API lets your app send a task to a model and get a response.

This is a different path from using CLI. CLI runs the agent team for you. With an API, you build the app around the model.

Keep API keys private.

An API key lets your app use a service. Keep it on your server, not in public browser code.

Keep full keys out of links, screenshots and shared files. Anyone with a key may be able to use its access.

Connect your app.

Connection settings and tested examples will be published here when API access opens.

What streaming means.

Streaming shows a response as it arrives. Your app can display the first part while the model writes the rest.

Show when the response is still loading. If it stops early, tell the user what happened.

Understand usage limits.

A request is one call to a service. Tokens are small pieces of text that a model reads or writes. They are different units.

Models and services have limits. Check those limits before sending a large task or many requests. AmirahCo’s models and limits will be listed when access opens.

When a request fails.

Check the API key, model name and usage limits. If the service is busy, wait before trying again.

Tell the user what they can do next. Keep error records useful, but don’t save full keys or private input in them.

Data and support.

Before connecting a service, check how it handles the data you send. Use only the data your task needs.

Read AI and your data. For questions, contact info@amirahco.us. Don’t include API keys or passwords.