forge

Automate development tasks with autonomous agents and external verification.

Updated Feb 4, 2026
One-click install
npx skills add https://github.com/vieko/forge --skill forge-vieko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: forge
Source: https://github.com/vieko/forge/tree/main/skills/forge
Command: npx skills add https://github.com/vieko/forge --skill forge-vieko

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates complex, multi-step development tasks by delegating them to an autonomous agent that builds and verifies code, ensuring outcomes are met rather than prescribing procedures.

Core Features & Use Cases

  • Outcome-Driven Development: Define desired outcomes in spec files, and let the agent implement them.
  • Automated Verification: Forge automatically detects project types, runs builds/tests, and feeds errors back to the agent for correction.
  • Use Case: Delegate the implementation of a new feature, including writing tests and ensuring it compiles, to Forge by providing a spec file describing the desired outcome.

Quick Start

Use the forge skill to run the spec file named 'implement-feature.md' with the description 'implement the new user authentication flow'.

Frequently Asked Questions about forge

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

FAQPage Schema
How do I automate complex development tasks using autonomous agents?

Automate complex development tasks by delegating them to an autonomous agent that implements code from spec files. The agent builds and verifies outcomes, feeding errors back for correction until specifications are met.

What is outcome-driven development and how does external verification work?

Outcome-driven development defines desired outcomes in spec files rather than prescribing procedures. External verification automatically detects project types, runs builds and tests, and feeds errors back to the agent for correction.

How do I delegate feature implementation and testing to an autonomous agent?

Delegate feature implementation by providing a spec file describing the desired outcome. The agent writes the code, runs tests, and automatically corrects errors against the defined specifications.

Can I use parallel processing and code auditing for multiple spec files?

Parallel processing and code auditing are supported for handling multiple spec files. The agent reviews git changes against defined specifications to ensure quality assurance across parallel tasks.

Does the agent automatically detect project types and run builds?

Automated verification automatically detects project types and runs builds and tests. Build and test errors are fed directly back to the agent for autonomous correction against the spec.

How do I define new specs from a feature description?

Define new specs by providing a text description of the desired outcome. The agent generates the spec file, implements the feature, and verifies the code compiles and passes tests.