forge-next

Execute one milestone, slice, or task and halt in step mode.

3|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/vh2224/forge-agent --skill forge-next
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: forge-next
Source: https://github.com/vh2224/forge-agent/tree/main/skills/forge-next
Command: npx skills add https://github.com/vh2224/forge-agent --skill forge-next

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables precise, incremental execution by running exactly one unit of work and then pausing, which is essential for careful validation within larger autonomous workflows.

Core Features & Use Cases

  • Single-step execution: runs one unit (milestone, slice, or task) and stops to allow verification.
  • Deterministic progress control: supports step/next semantics to isolate failures and assess outcomes.
  • Use Case: ideal when integrating with planners and executors to validate a concrete unit before proceeding to the next.

Quick Start

Run /forge-next to execute exactly one unit in step mode.

Frequently Asked Questions about forge-next

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

FAQPage Schema
How do I execute a single unit of work and pause for verification in an autonomous workflow?

To execute a single unit of work and pause, use a step mode execution tool that runs exactly one task, milestone, or slice before halting. This allows precise incremental validation within larger automated workflows.

What is step mode execution for milestone and slice workflows?

Step mode execution is an incremental progress control mechanism that processes one defined unit of work at a time. It isolates outcomes by stopping after each slice or milestone, ensuring deterministic validation before proceeding.

How do I isolate failures when running tasks in an automated execution pipeline?

You can isolate failures by using deterministic step or next semantics to execute a single task. By pausing after each unit runs, you can assess the outcome and accurately pinpoint where the execution pipeline fails.

Can I validate a specific task before proceeding to the next in an autonomous workflow?

Yes, you can validate a specific task by running the execution in single-step mode. This processes one unit of work and halts, allowing you to verify the concrete outcome before moving to the next automated step.

Do I need specific dependencies to run single-step execution for incremental verification?

No specific dependencies are required to run single-step execution for incremental verification. The execution operates independently, adhering to frontmatter-defined unit behavior to halt after processing a single slice or task.

When should I use single-unit execution instead of continuous workflow automation?

You should use single-unit execution instead of continuous automation when careful validation is essential. It is ideal for isolating failures and assessing outcomes in milestone or task workflows where verifying one step before proceeding is critical.