ARchetipo Beta

Welcome to the ARchetipo beta

Thank you for choosing to join the ARchetipo beta. Your contribution will help us see the product at work in real projects, teams and contexts, exactly where the most useful feedback emerges to make it clearer, stronger and more reliable.

ARchetipo is a spec-driven workflow for AI coding agents. Instead of treating every request as an isolated prompt, it introduces a shared method, specialized roles and persistent artifacts that accompany the whole development path: from the initial vision to the backlog, from specs to the technical plan, through implementation, testing and review.

Its main characteristics include high token efficiency thanks to a CLI-driven approach, the ability to save artifacts through dedicated connectors and a local board to track work status. It also adapts to the language of the conversation, so backlogs, plans and documents remain readable and consistent in real team flows.

Essential resources

• Technical documentation
• Product page
• GitHub repository
• Discord server

What to test

The most useful tests for us are the ones that put ARchetipo under pressure in a real context: different environments, different projects and different workflows.

Setup and installation

The CLI supports global installation and initialization across multiple tools. With the archetipo init command you can install the skills for Claude Code, Codex, Gemini CLI, OpenCode, GitHub Copilot and Pi, as well as create .archetipo/config.yaml and .archetipo/shared-runtime.md. After installation, you can run archetipo version to verify that the CLI is available and check the installed version.

So if you want to contribute to ARchetipo development, try the installation on macOS, Linux or Windows. Check whether npm install -g @techreloaded/archetipo is straightforward or requires workarounds. Check whether archetipo init clearly explains the choice between tool and connector.

Complete workflow

ARchetipo's value emerges when the whole flow stays coherent from beginning to end. The most useful test is to follow a real need through the whole path, checking whether each step prepares the next one well and whether the artifacts remain understandable after multiple iterations.

For this reason, start from a concrete project goal, not from a prompt disconnected from the context. Check whether discovery, backlog, spec, acceptance criteria, tasks and review really tell the same story. Observe whether the technical plan produced by archetipo-plan is precise enough to guide decisions, dependencies and work order. Check whether archetipo-implement stays coherent with the plan, updating code, tests and final status in a readable way.

Connectors and board

ARchetipo works on project artifacts, such as backlogs, specifications and plans, that can be connected to different tools through the connector. The connector decides where these artifacts are read, saved and updated during the workflow.

Today ARchetipo supports the file and github connectors, plus a local board available through archetipo view. If you want to understand this part in more detail, see the technical documentation about connectors. During the beta, we want to verify whether the workflow remains robust when artifacts move.

Language and output

ARchetipo is designed to adapt to the language of the conversation. If you work in Italian or alternate between Italian and English, we want to know whether the artifacts remain consistent and whether the generated text is suitable for a team to read, not only for a model.

• Check whether backlogs, plans and notes remain in the right language.
• Check whether technical names and states remain clear.
• Report outputs that are too verbose, too vague or difficult to review quickly.

What we expect from beta testers

We do not need a "clean" test. We need a real test. The best feedback arrives when ARchetipo enters a real context and meets real constraints.

• Use a concrete project. Even better if you use it on a flow that already has context, constraints and steps to coordinate.
• Tell us where the method really helps. Not only where it fails, but also where it makes work more ordered or faster.
• Send reproducible feedback. The clearer the context, the faster we can fix issues.
• If you liked ARchetipo, leave a star on the GitHub repository.

In practice, it is very useful for us if you share this information:

• AI tool used: Claude Code / Codex / Gemini CLI / OpenCode / Copilot / Pi
• Operating system: Windows / macOS / Linux
• Selected connector: file / github
• Command executed: archetipo init | archetipo view | /archetipo-plan US-001 | ...
• What you expected:
• What actually happened:
• Any screenshot / log / file involved:

How to send feedback

The most direct channel is our Discord server. It is the right place for quick doubts, small setup friction, screenshots and immediate observations while you are testing.

We at Tech Reloaded will be available for voice chats, immediate support sessions, direct chats and feedback collection during the beta.

If you prefer, you can also write to us on LinkedIn or through the form at the bottom of the page.

Quick checklist to get started

• Open the technical documentation.
• Install the CLI.
• Initialize a project with archetipo init.
• Create or choose a real spec to work on.
• Send us feedback with context, command and result.

Learn more

The technical documentation covers installation, prerequisites, quickstart, glossary, CLI, connector and configuration for using ARchetipo with an AI coding agent.

The product page presents ARchetipo as an AI team by your side, from idea to code: a complete product development process with specialized roles and persistent artifacts.