intent-frame

Construct intent frames with objectives, health metrics, guardrails, and decision boundaries.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/tgsatwit/agent-delivery-pub --skill intent-frame
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: intent-frame
Source: https://github.com/tgsatwit/agent-delivery-pub/tree/main/.claude/skills/intent-frame
Command: npx skills add https://github.com/tgsatwit/agent-delivery-pub --skill intent-frame

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of creating well-defined and comprehensive intent frames for new agents, specifications, features, or initiatives, ensuring clarity and alignment.

Core Features & Use Cases

  • Structured Intent Definition: Guides users through defining objectives, health metrics, guardrails, and decision boundaries.
  • Alignment and Validation: Incorporates OKR alignment and the "Klarna Test" for outcome validation.
  • Use Case: When launching a new AI agent for customer support, use this Skill to meticulously define its goals, how its success will be measured, what actions it must avoid, and the trade-offs it can make.

Quick Start

Use the intent-frame skill to build a new intent frame for a customer support agent.

Frequently Asked Questions about intent-frame

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

FAQPage Schema
How do I define guardrails and decision boundaries for a new AI agent?

An intent frame is a structured specification defining an agent's objectives, health metrics, guardrails, and decision boundaries. It is needed when launching new features or initiatives to ensure clarity, alignment, and validated outcomes.

How do I ensure my initiative's objectives align with OKRs and validate outcomes?

Ensure objective alignment and validate outcomes by building a structured intent frame that incorporates OKR checks and applies the Klarna Test. This process validates that your specified health metrics and boundaries meet intended goals.

What is the best way to structure specifications for a customer support agent?

The best way to structure agent specifications is to interactively build an intent frame. This method requires structured input for each component, meticulously defining goals, success measurements, avoided actions, and trade-offs.

Do I need structured input for every component when building an agent framework?

Yes, structured input is required for each component of the intent frame. The framework interactively constructs objectives, health metrics, guardrails, and decision boundaries, ensuring no specification details are missed during alignment.

Can I use intent frames to manage trade-offs for new product features?

Yes, you can use intent frames to manage trade-offs for new product features. The framework defines decision boundaries and guardrails, detailing exactly what actions the feature must avoid and what compromises it can make.