ICE

Structure execution traces with Intent, Condition, and Eval sections.

Updated Jan 21, 2026
One-click install
npx skills add https://github.com/terateams/TAES --skill ice
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ICE
Source: https://github.com/terateams/TAES/tree/main/.github/skills/ICE
Command: npx skills add https://github.com/terateams/TAES --skill ice

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a standardized framework for creating verifiable, auditable execution traces by structuring the three essential elements: Intent, Condition, and Eval, ensuring every action is traceable from planning to completion.

Core Features & Use Cases

  • Intent-driven task definition: clearly specify mission goals, success criteria, and rationale to align collaborators.
  • Condition-based gating: define readiness, constraints, and dependencies to gate progress and maintain control.
  • Eval-based evidence: capture phase-specific results and artifacts to build an auditable history for reviews, audits, and handover.

Quick Start

To begin, define an Intent with Mission, Value, and Success, then specify Conditions and an Eval plan to track progress through the project lifecycle.

Frequently Asked Questions about ICE

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

FAQPage Schema
How do I create auditable execution traces for manual workflows and AI collaboration?

Auditable execution traces are created by structuring tasks into Intent, Condition, and Eval sections. This standardizes evidence collection and documents lifecycle milestones from kickoff to handover for verifiable tracking.

What is the ICE framework for mission step traceability?

The ICE framework is a standardized structure for execution tracking that defines Intent, Condition, and Eval. It enforces explicit sections to ensure every action and outcome is traceable from planning to completion.

How do I structure intent and conditions to gate task progress?

To gate task progress, define an Intent with Mission, Value, and Success criteria, then specify Conditions for readiness, constraints, and dependencies. This maintains control and aligns collaborators before execution proceeds.

Can I use this structured execution tracking for manual workflows?

Yes, this structured execution tracking supports use cases across manual workflows and AI collaboration. It captures phase-specific results and artifacts to build an auditable history for reviews, audits, and handover.

Why use condition-based gating for execution tracking?

Condition-based gating is used to define readiness, constraints, and dependencies before proceeding. It enforces explicit checks to maintain control, ensuring traceable decisions are made only when defined conditions are met.

What are the limitations of manual execution tracking without structured eval evidence?

Without structured eval evidence, manual execution tracking lacks standardized artifact collection and auditable history. This makes it difficult to verify outcomes, align collaborators, or pass reviews during project handover.