THE AI AGENT GLOSSARY
Half the words around AI sound like jargon built to keep you out. Here are the terms that actually matter — each in one plain sentence, linked to where it lives in the Engine.
An AI that doesn't just answer — it acts: takes a goal, decides the steps, uses tools, and keeps working until the job is done.
Browse the agent directory →The software wrapped around the model that runs the actual work: reads the output, runs tools, feeds results back, retries failures, and decides when to stop.
Read: What is an agent harness? →A layer above the harness that coordinates several agents at once, handing tasks between them.
Read: What is an agent orchestrator? →Plain markdown files you hand your AI that tell it who it is, what the rules are, and how you want the work done. The highest-leverage thing you control.
See sample Context Files →The craft of deciding what goes into the AI's limited working memory each turn, and what to leave out.
Read: Prompt vs context vs harness →An open standard that lets an AI plug into outside tools and data — your email, files, calendar — through one common connector.
MCP starter packs by niche →Reusable, packaged instructions (often a folder of markdown and files) that teach an agent one task well, loaded only when needed.
Skills starter packs →A pre-packaged, self-contained workspace that gives the AI a clean, consistent place to run with the right tools already installed.
Run the cloud-install wizard →The cycle an agent repeats: think, act (use a tool), read the result, decide the next step — until the job is done.
Which AI layer do you need? →The moment the AI stops writing text and actually does something: searches the web, sends an email, edits a file.
The full glossary article →EVERY TERM LOOPS BACK TO ONE LEVER
The harness, the tools, the loop run in the background. The markdown you hand your AI is what makes the output yours. The $1 Starter Kit generates your first Context Files.
GET THE $1 STARTER KIT →