Glossary of AI agents and the agent-ready web
Plain-English definitions of the terms behind AI agents on websites, from llms.txt and WebMCP to RAG, grounding and lead qualification.
50 terms
Terms starting with A
Agent readiness
Agent readiness is how well a website can be found, read and used by AI agents acting for people, from assistants that research vendors to browsers that fill in forms.
Agentic browser
An agentic browser is a web browser with a built-in AI agent that can read pages and act on them for its user: clicking, filling in forms and finishing tasks.
Agentic commerce
Agentic commerce is buying and selling where an AI agent finds, compares and purchases products or services for a person, within limits that person sets.
AI agent
An AI agent is software that uses a language model to reach a goal by choosing its own next steps, calling tools and reacting to the results.
AI crawler
An AI crawler is a bot run by an AI company that fetches web pages to train models, build an AI search index or answer a user's question in real time.
AI Overview
An AI Overview is a short AI-written summary that Google Search shows with its results, with links to the web pages it drew on.
AI SDR
An AI SDR is an AI agent that does the early work of a sales development rep: engaging inbound leads, asking qualifying questions and booking meetings.
Answer engine
An answer engine is a search tool that replies to a question with a written answer and links to its sources, instead of a list of results to sift through.
Answer engine optimization (AEO)
Answer engine optimization (AEO) is structuring content so search engines and AI assistants can lift a direct, accurate answer to a question from your page.
Terms starting with C
Chunking
Chunking is splitting long documents into smaller passages before indexing, so an AI system can retrieve just the part of a page that answers a question.
Citation
A citation in an AI answer is a link or reference showing which source a statement came from, so the reader can check it and read further.
Content Security Policy (CSP)
A Content Security Policy (CSP) is a set of rules a site sends to the browser listing where the page may load scripts and other resources from and send data to.
Context window
A context window is the maximum amount of text, measured in tokens, that a language model can take into account at once, including its instructions and reply.
Conversion rate
Conversion rate is the share of visitors or sessions that complete a goal you care about, such as a demo request, sign-up or purchase, over a set period.
Conversion rate optimization (CRO)
Conversion rate optimization (CRO) is the practice of raising the share of visitors who complete a goal by finding friction and testing changes that remove it.
CRM (customer relationship management)
A CRM (customer relationship management system) is software that keeps one record per contact, company and deal, with the history of every interaction, so a team can see where each relationship stands.
Terms starting with E
Embeddings
Embeddings are lists of numbers that represent the meaning of a piece of text, so software can find related content by measuring how close two lists are.
EU AI Act
The EU AI Act is the European Union's law on artificial intelligence.
Terms starting with G
GDPR (General Data Protection Regulation)
The GDPR is the EU's data protection law.
Generative engine optimization (GEO)
Generative engine optimization (GEO) is the practice of making content more likely to be found, used and cited by AI systems that write answers to questions.
Grounding
Grounding means tying an AI model's answer to specific, verifiable sources, such as your own pages or a search index, so each claim can be checked.
Guardrails
Guardrails are checks placed around an AI model that limit what it accepts, says and does, so its behavior stays inside rules the business has set.
Terms starting with H
Hallucination
An AI hallucination is a fluent, confident answer from a language model that is false or unsupported, such as an invented price, policy or citation.
Human handoff
Human handoff is when an AI agent passes a conversation to a person, with the transcript and key details, because the visitor asked or the AI reached a limit.
Terms starting with K
Knowledge base
A knowledge base is the curated set of content an AI agent is allowed to answer from, such as web pages, help articles, PDFs and FAQs, kept indexed and current.
Terms starting with L
Large language model (LLM)
A large language model (LLM) is an AI model trained on vast amounts of text to predict the next token, which lets it write, summarize, translate and answer.
Lead qualification
Lead qualification is deciding whether a potential buyer fits your offer and is ready to talk, so sales time goes to the leads most likely to become customers.
Lead scoring
Lead scoring gives each potential buyer a number based on who they are and what they have done, so sales and marketing know whom to contact first.
llms-full.txt
llms-full.txt is an informal convention for publishing a site's documentation as one Markdown file, so AI tools can load all of it in a single request.
llms.txt
llms.txt is a proposed Markdown file at a website's root that gives AI tools a short summary of the site and links to the pages that matter most.
Terms starting with M
Meeting booking
Meeting booking lets a website visitor pick a time on your team's calendar and confirm a call on the spot, without emails going back and forth.
Model Context Protocol (MCP)
The Model Context Protocol (MCP) is an open standard for connecting AI applications to outside data, tools and workflows through one common interface.
Terms starting with O
OAuth
OAuth is a standard that lets one application use your account in another with limited, revocable permission, without you handing over your password.
Terms starting with P
Prompt injection
Prompt injection is when text an AI model reads, typed by a user or hidden in a page or file, overrides its instructions and changes what it says or does.
Terms starting with R
Reranking
Reranking is a second search step in which a model reorders a shortlist of retrieved passages by how well each one actually answers the user's question.
Retrieval
Retrieval is the step where an AI system searches a document collection for the passages most likely to answer a question, before a model writes the answer.
Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG) is a technique where an AI system first looks up relevant documents, then writes its answer using what it found.
Reverse trial
A reverse trial starts new users on a product's paid features for a limited time, then moves them to a free plan instead of cutting off access.
robots.txt
robots.txt is a plain-text file at the root of a website that tells crawlers which paths they may fetch, following the Robots Exclusion Protocol (RFC 9309).
Terms starting with S
Search intent
Search intent is the goal behind a query, such as learning something, comparing options, finding a site or taking an action, which shapes what a good result is.
Semantic search
Semantic search finds results by what a query means rather than the exact words it contains, so different wording can still lead to the right page.
Structured data
Structured data is machine-readable markup, usually schema.org terms written as JSON-LD, that states what a page describes: a product, company or article.
System prompt
A system prompt is the set of instructions an application gives a language model before a conversation starts: its role, rules, tone and what it must not do.
Terms starting with T
Tokens
Tokens are the small units of text, often a word or part of a word, that language models read and write, and the unit in which AI usage is usually priced.
Terms starting with V
Vector database
A vector database stores embeddings and quickly finds the ones closest to a query, which lets AI systems search content by meaning instead of exact words.
Vector embedding
A vector embedding is one fixed-length list of numbers that a model produces for a piece of content, placing it near other content with a similar meaning.
Terms starting with W
Webhook
A webhook is an HTTP request one application sends to a URL you choose as soon as a specific event happens, carrying the data about that event.
WebMCP
WebMCP is a proposed web standard that lets a website register its own actions as structured tools that AI agents in the browser can discover and call.
Website widget
A website widget is a ready-made piece of functionality, such as a chat box or booking calendar, that a site adds by pasting a short code snippet.
Terms starting with Z
Zero-click search
Zero-click search is a search that ends without a click to any website, because the answer was shown on the results page or the searcher gave up.