OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Coding Agents

Armada

Armada is a JavaScript command-line developer tool that coordinates eight specialist AI engineering roles around software changes in repositories. It combines contract creation, dependency-aware dispatch, isolated Git worktrees, evidence gates and pull-request delivery into a loop for planning, building, testing and reviewing features.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Armada — OpenViST project profile in Coding Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Armada when a repository change needs structured requirements, multiple specialist roles, explicit verification and isolated parallel feature work. It is also suited to developers who want the workflow state and evidence gates recorded rather than relying on a single unsupervised coding session.

Who it is for

  • Developers working on existing software repositories
  • Teams coordinating multiple AI-assisted coding tasks
  • Engineers who want contract-driven requirements and evidence before workflow transitions
  • Users operating a self-hosted CLI-based development process

Documented uses

  • Turn a repository into a coordinated AI engineering workflow
  • Convert a software goal into a contract with phases and measurable success criteria
  • Delegate backend, frontend, QA, security, documentation and review work
  • Run independent feature phases in parallel with isolated file scope
  • Require test, screenshot or file-and-line evidence before advancing
Before you choose

Requirements & limits

The documented setup requires Node.js 22 or later and an authenticated opencode installation. Armada is distributed as the npm package @rafmacalaba/armada and the repository is public under the MIT license.

  • The workflow depends on an authenticated opencode installation in addition to Node.js and Armada.
  • Requirements are formalized in armada/REQUIREMENTS.md before the fleet runs autonomously.
  • Independent phases can run in parallel, while dependent phases are serialized.
  • Each feature uses a separate Git worktree, branch and state, with delivery ending in a pull request.
  • Evidence gates accept items such as passing test runs, screenshots or file-and-line citations.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Armada in an existing repository

Install Armada and scaffold its AI engineering team in an existing repository.

Before you start

  • Node.js 22 or later
  • An authenticated opencode installation
  • An existing software repository
  • npm or npx

Steps

  1. Install the package globally with npm install -g @rafamacalaba/armada, or use the documented npx package command.

  2. Change into the existing repository.

  3. Run armada init to detect the stack and scaffold the team in place.

  4. Run armada doctor to confirm the documented environment is ready.

Expected outcome

The repository is initialized for the Armada workflow and the environment check has been requested.

Guide limitations

  • This recipe documents setup and environment checking only; it does not verify that a feature run or pull request will complete successfully.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
4 October 2026
Original website title
rafmacalaba/armada · Official homepage title · retained in its source language
How we review sourcesSuggest a correction