goal

Drive a single engineering objective through verified completion and pull request handoff.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/ArthTech-AI/Marketplace-extensions --skill goal-arthtech-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: goal
Source: https://github.com/ArthTech-AI/Marketplace-extensions/tree/main/skills/goal
Command: npx skills add https://github.com/ArthTech-AI/Marketplace-extensions --skill goal-arthtech-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you pursue one open-ended objective without losing momentum, by turning a broad goal into a focused, evidence-backed execution loop that stops at pull request review.

Core Features & Use Cases

  • Single-objective autonomy: Takes one freeform outcome and drives the next concrete action instead of juggling a backlog.
  • Stateful progress tracking: Stores goal state, subtasks, clarifications, evidence, and connections so work can pause and resume cleanly.
  • Verification-first execution: Requires linting, testing, and other proof before marking work complete.
  • Use case: Use it when you want an AI assistant to improve performance, fix a bug, or ship a scoped change while keeping the process disciplined and auditable.

Quick Start

Use the goal skill to work toward your objective by stating the outcome you want and letting the loop plan, execute, and verify each step until it is ready for pull request review.

Frequently Asked Questions about goal

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

FAQPage Schema
How do I automate an open-ended engineering task with state tracking and verification?▼

Automate an open-ended engineering task by using an autonomous loop that plans, executes, and verifies each step. It tracks goal state and subtasks, requires linting and testing as proof, and stops at pull request review for disciplined progress.

What is verification-first execution for autonomous coding workflows?▼

Verification-first execution requires mandatory linting, testing, and other proof before marking work complete. This ensures autonomous engineering loops produce evidence-backed progress rather than unverified code changes.

Can I pause and resume a long-running objective without losing progress?▼

Pause and resume long-running objectives cleanly through stateful progress tracking. The loop stores goal state, subtasks, clarifications, and evidence so work can stop and restart without losing momentum or context.

What's the best way to turn a broad engineering goal into a pull request?▼

Turn a broad goal into a pull request by stating the desired outcome to drive an autonomous execution loop. It plans concrete actions, verifies progress through testing, and hands off the scoped change for review.

Does this goal execution approach work for fixing bugs and shipping scoped changes?▼

This goal execution approach works for fixing bugs and shipping scoped changes by driving one open-ended objective at a time. It maintains auditable progress through state tracking and mandatory verification before handoff.