<Agentic WorkFlow> Glossary

AI agent

AI agents are artificial intelligence systems capable of responding to requests, making decisions, and performing real-world tasks for users. They use large language models (LLMs) to interpret user input and make decisions about how to best process requests using the information and resources they have available.

AI chain

AI chains allow you to interact with large language models (LLMs) and other resources in sequences of calls to components. AI chains in Agentic WorkFlow don’t use persistent memory, so you can’t use them to reference previous context (use AI agents for this).

AI completion

Completions are the responses generated by a model like GPT.

AI embedding

Embeddings are numerical representations of data using vectors. They’re used by AI to interpret complex data and relationships by mapping values across many dimensions. Vector databases, or vector stores, are databases designed to store and access embeddings.

AI groundedness

In AI, and specifically in retrieval-augmented generation (RAG) contexts, groundedness and ungroundedness are measures of how much a model’s responses accurately reflect source information. The model uses its source documents to generate grounded responses, while ungrounded responses involve speculation or hallucination unsupported by those same sources.

AI hallucination

Hallucination in AI is when an LLM (large language model) mistakenly perceives patterns or objects that don’t exist.

AI reranking

Reranking is a technique that refines the order of a list of candidate documents to improve the relevance of search results. Retrieval-Augmented Generation (RAG) and other applications use reranking to prioritize the most relevant information for generation or downstream tasks.

AI memory

In an AI context, memory allows AI tools to persist message context across interactions. This allows you to have a continuing conversations with AI agents, for example, without submitting ongoing context with each message. In Agentic WorkFlow, AI agent nodes can use memory, but AI chains can’t.

AI retrieval-augmented generation (RAG)

Retrieval-augmented generation, or RAG, is a technique for providing LLMs access to new information from external sources to improve AI responses. RAG systems retrieve relevant documents to ground responses in up-to-date, domain-specific, or proprietary knowledge to supplement their original training data. RAG systems often rely on vector stores to manage and search this external data efficiently.

AI tool

In an AI context, a tool is an browser extension resource that the AI can refer to for specific information or functionality when responding to a request. The AI model can use a tool to interact with external systems or complete specific, focused tasks.

AI vector store

A vector store, or vector database, stores mathematical representations of information. Use with embeddings and retrievers to create a database that your AI can access when answering questions.

API

APIs, or application programming interfaces, offer programmatic access to a service’s data and functionality. APIs make it easier for software to interact with external systems. They’re often offered as an alternative to traditional user-focused interfaces accessed through web browsers or UI.

Browser Context Manipulation

The ability to interact with and extract data from web pages within the browser environment. This includes extracting text, HTML, links, images, and other content from the current web page or selected elements.

Browser Extension Node

Specialized workflow components that can interact with browser context, such as extracting selected text, collecting all links, or gathering images from web pages.

Canvas

The main interface for building workflows in Agentic WorkFlow’s browser-based editor. You use the canvas to add and connect nodes to compose browser automation workflows.

cluster node (Agentic WorkFlow)

In Agentic WorkFlow, cluster nodes are groups of nodes that work together to provide functionality in a workflow. They consist of a root node and one or more sub nodes that extend the node’s functionality.

Credential

In Agentic WorkFlow, credentials store authentication information to connect with external apps and services from within the browser environment. Browser security limitations may restrict certain types of authentication methods.

Data Pinning

Data pinning allows you to temporarily freeze the output data of a node during workflow development. This is particularly useful when working with browser context data, allowing you to develop workflows with predictable web page data without repeatedly extracting content from web pages.

Editor

The Agentic WorkFlow editor runs directly in your browser and allows you to create and manage browser-based workflows. The main area is the canvas, where you can compose workflows by adding, configuring, and connecting nodes. The interface provides access to browser extension nodes, credentials, templates, and execution history.

Entitlement (Agentic WorkFlow)

In Agentic WorkFlow, entitlements grant browser extension instances access to plan-restricted features for a specific period of time.

Floating entitlements are a pool of entitlements that you can distribute among various Agentic WorkFlow instances. You can re-assign a floating entitlement to transfer its access to a different Agentic WorkFlow instance.

Evaluation (Agentic WorkFlow)

In Agentic WorkFlow, evaluation allows you to tag and organize execution history and compare it against new executions. You can use this to understand how your workflow performs over time as you make changes. This is particularly useful while developing AI-centered workflows in the browser environment.

Expression

In Agentic WorkFlow, expressions allow you to populate node parameters dynamically by executing JavaScript code within the browser environment. You can use expressions to process browser context data, manipulate extracted text, or create dynamic values based on web page content.

LangChain

LangChain is an AI-development framework used to work with large language models (LLMs). LangChain provides a standardized system for working with a wide variety of models and other resources and linking different components together to build complex applications.

Large language model (LLM)

Large language models, or LLMs, are AI machine learning models designed to excel in natural language processing (NLP) tasks. They’re built by training on large amounts of data to develop probabilistic models of language and other data.

Node

In Agentic WorkFlow, nodes are individual components that you compose to create browser-based workflows. Nodes can extract data from web pages, process browser context information, define flow control logic, and connect with external services within browser security constraints.

Project (Agentic WorkFlow)

Agentic WorkFlow projects allow you to separate workflows, variables, and credentials into separate groups for easier management. Projects make it easier for teams to collaborate by sharing and compartmentalizing related resources.

Root node (Agentic WorkFlow)

Each Agentic WorkFlow cluster node contains a single root node that defines the main functionality of the cluster. One or more sub nodes attach to the root node to extend its functionality.

Sub node (Agentic WorkFlow)

Agentic WorkFlow cluster nodes consist of one or more sub nodes connected to a root node. Sub nodes extend the functionality of the root node, providing access to specific services or resources or offering specific types of dedicated processing, like calculator functionality, for example.

Template

Agentic WorkFlow templates are pre-built browser automation workflows designed by the community. These templates demonstrate common browser context manipulation patterns and can be imported and customized for your specific web automation needs.

Trigger Node

A trigger node is a special node responsible for executing browser-based workflows in response to certain conditions. In the browser environment, triggers can respond to page events, user interactions, or manual execution.

Workflow

An Agentic WorkFlow workflow is a collection of nodes that automate browser-based processes. Workflows can extract data from web pages, process content with AI, and perform complex browser automation tasks.

Web Page Context

The current state and content of a web page that can be accessed by browser extension nodes. This includes visible text, HTML structure, links, images, and user selections.

Content Security Policy (CSP)

Browser security restrictions that may limit certain workflow operations. Agentic WorkFlow respects CSP limitations and provides guidance on working within browser security constraints.

DOM Manipulation

The ability to interact with and extract data from the Document Object Model (DOM) of web pages. Browser extension nodes use DOM manipulation to access page content and structure.

Cross-Origin Restrictions

Browser security limitations that prevent workflows from accessing content from different domains. Understanding these restrictions is important for designing effective browser automation workflows.