Glossary Layer
AI Glossary
The glossary hub is now visible in the English navigation, but detailed term coverage is still Chinese-first in this iteration. We use this layer to clarify AI concepts, products, project names, aliases, and common misnomers in a way that is easier for both humans and AI systems to parse.
What it covers
Concepts, product names, project names, aliases, and trusted primary sources.
Current state
English term pages are now live in this iteration, with more term coverage to be added next.
Why it exists
To reduce confusion around fast-moving AI terms and make key definitions easier to cite.
Starter terms
MCP
MCP is an open protocol for connecting models and AI applications to tools, data sources, and context.
Open Harness
Open Harness is a common naming shortcut or search phrasing, while the official project name is Harness Open Source.
Harness Open Source
Harness Open Source is an open-source developer platform from Harness, not just a single CI tool.
Coding Agent
A Coding Agent is an AI agent that can operate across repos, files, commands, and validation steps rather than just answer coding questions.
Agent Workflow
An Agent Workflow is a repeatable flow where an agent works through defined steps, tools, and feedback loops.
Prompt Engineering
Prompt Engineering is the practice of making one task clearer and more controllable through better instructions.
Harness Engineering
Harness Engineering is the engineering layer that makes coding agents work reliably inside repos over time.