Glossary
Core concepts and terminology for Intent Engineering and AI-native product development.
Core Concepts
Agent-Ready Spec
A specification structured so that an AI coding agent can execute it without follow-up questions, containing clear objectives, measurable outcomes, edge cases, and verification criteria.
AI Coding Agent
An AI system that can autonomously write, modify, and test code based on natural language instructions or structured specifications.
Intent Engineering
The discipline of specifying what you want built precisely enough that an AI agent can execute it without guessing.
Intent Layer
The system layer between user research and code execution that translates raw friction signals into structured, agent-ready specifications.
IntentSpec
A structured specification format with six parts — Objective, Outcomes, Evidence, Constraints, Edge Cases, and Verification — designed for both human review and AI agent execution.
Prompt Engineering
The practice of crafting and optimizing natural language inputs to get better outputs from AI models.
Methodology
Friction Signal
A discrete piece of evidence — a support ticket, user quote, drop-off metric, or feature request — that indicates where users struggle with a product.
Vibe Coding
Building software by describing what you want to an AI coding agent in natural language, iterating through conversation rather than writing code directly.
Product Strategy
North Star
A single measurable signal that tells you whether your product is moving toward its vision — the quantitative counterpart to qualitative direction.
Product Vision
A concise statement of where your product is headed and why — the strategic direction that shapes what gets built and what gets ignored.
Put these concepts into practice.
Pathmode turns Intent Engineering principles into a working platform — from evidence collection to agent-ready specs.
Get Started with Pathmode