goalpro

Convert vague requests into goal contracts and loop prompts for AI agents.

160|49|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/KimYx0207/Kim_Service --skill goalpro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: goalpro
Source: https://github.com/KimYx0207/Kim_Service/tree/main/skills/goalpro
Command: npx skills add https://github.com/KimYx0207/Kim_Service --skill goalpro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

GoalPro helps you turn vague, strategic, multi-step, or evidence-poor requests into clear task contracts that an AI agent can execute, verify, pause on safely, and continue improving after delivery. It is designed for cases where users want better goals, clearer success criteria, stronger evidence standards, or a safer handoff to Codex or Claude Code.

Core Features & Use Cases

  • Goal Prompt generation: Reframes fuzzy intent into an actionable goal contract with scope, boundaries, verification, and stop conditions.
  • Loop Prompt generation: Creates a post-delivery continuation prompt with timing, state carryover, evidence review, gap diagnosis, and next-step packaging.
  • Workflow judgment: Detects repeated work, ongoing operations, publishing, monitoring, or automation-style requests and folds only the necessary trigger, checkpoint, and brief information into the two prompt outputs.
  • Research and repair support: Handles strategic research tasks, big refactors, and “this is too complex” corrections by adding evidence requirements, inventory, and safer execution policies.

Quick Start

Ask GoalPro to rewrite your fuzzy request into a copyable Goal Prompt and Loop Prompt for an AI agent.

Frequently Asked Questions about goalpro

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

FAQPage Schema
How do I turn vague prompts into executable goal contracts for AI agents?

Goal contracts reframe fuzzy intent into actionable prompts with defined scope, success criteria, evidence standards, boundaries, and pause conditions for AI agents.

What is a loop prompt and how does it work for iterative agent workflows?

A loop prompt is a post-delivery continuation prompt that packages timing, state carryover, evidence review, gap diagnosis, and next-step packaging to drive iterative improvement in agent workflows.

How do I write task contracts with evidence standards for Codex and Claude Code?

Writing task contracts with evidence standards involves adding intent expansion, verification rules, inventory, and safer execution policies to ensure AI agents like Codex and Claude Code execute safely.

Can I use prompt engineering to scope complex refactoring tasks for AI agents?

Prompt engineering can scope complex refactoring tasks by transforming multi-step, evidence-poor requests into structured goal contracts with inventory, boundaries, and stop conditions for safer execution.

Does this approach support ongoing monitoring and automation-style agent requests?

Ongoing monitoring and automation-style requests are supported by folding necessary triggers, checkpoints, and brief information into the goal prompt and loop prompt outputs.

What are the limitations of using goal contracts for multi-step AI agent tasks?

Goal contracts require complete intent expansion, explicit pause conditions, and continuation state formatting, adding overhead that may not suit simple, single-step prompt engineering tasks.