Forge

Coordinate eight autonomous agents to verify Gherkin specs and run E2E tests.

9|2|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/ikennaokpala/forge --skill forge-ikennaokpala
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Forge
Source: https://github.com/ikennaokpala/forge/tree/main
Command: npx skills add https://github.com/ikennaokpala/forge --skill forge-ikennaokpala

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Forge automates quality assurance by coordinating an autonomous swarm of agents to verify behavior and integrity of production-ready software.

Core Features & Use Cases

  • Autonomous multi-agent QA: 8 specialized agents working in parallel to verify Gherkin specs, run E2E tests, perform defect analysis, and apply fixes.
  • 7 Quality Gates: Functional, Behavioral, Coverage, Security, Accessibility, Resilience, Contract—ensuring end-to-end quality.
  • ADR-driven architecture: Architecture Decision Records to document dependencies and decisions; continuous learning via the Learning Optimizer.
  • No mocking rule: Tests run against the real backend to catch integration issues.
  • Deployment architecture adaptability: Works with monoliths, microservices, modular monoliths, and monorepos.

Quick Start

Copy the Forge skill into your Claude Code skills directory and run the autonomous workflow against your target context with the appropriate CLI switch.

Frequently Asked Questions about Forge

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How does autonomous E2E testing work with Gherkin specifications?

Automated defect prediction for microservices uses an autonomous agent swarm to analyze failures and apply confidence-tiered fixes. Forge coordinates 8 agents that run E2E tests against your live backend, performing defect analysis and reporting across 7 quality gates.

Can I run automated quality engineering tests against a live backend without mocking?

Yes, you can run automated quality engineering tests against a live backend without mocking. Forge enforces a strict no-mocking rule, executing E2E tests against the real backend to catch genuine integration issues across monoliths, microservices, and monorepos.

Do I need Architecture Decision Records to automate E2E testing for microservices?

Yes, you need Architecture Decision Records to automate E2E testing for microservices with Forge. ADRs document dependencies and architectural decisions, enabling the autonomous agent swarm to perform accurate context discovery and continuous learning via the Learning Optimizer.

What is the best way to automate defect prediction across monorepos and modular monoliths?

The best way to automate defect prediction across monorepos and modular monoliths is deploying an autonomous QE swarm. Forge adapts to various deployment architectures, using 8 specialized agents to perform defect analysis and enforce 7 quality gates including security, accessibility, and resilience.

How do I set up autonomous quality gates for behavioral and contract testing?

To set up autonomous quality gates for behavioral and contract testing, configure the autonomous workflow with the appropriate CLI switch. Forge requires a live backend, Gherkin specifications, ADRs, and configurable wiring to drive the 8 specialized agents across 7 quality gates.