Heidelberg AI Curriculum

Showing 1–40 of 47 terms matching “document”

Actions Concepts: Onyx's mechanism for letting an agent call external applications (with flexible authentication), so a chat can do things — look up or update a record — not just retrieve documents.
Add Source Interface: Button in the interface that lets you attach a new document or web page to the current notebook.
Agent Components Concepts Core: An AI system that does more than answer: it plans steps towards a goal, uses tools such as search or files, checks the result and carries on. Give it only the access the task needs.
Also known as: agent, AI agent
5 other definitions
Agent · Concepts
An AI process that Antigravity launches to carry out a task — it can read files, run code, browse the web, and make changes autonomously.
Agent · Concepts
A Dify application type where the AI can reason, decompose tasks, and call tools on its own without you having to pre-define every step.
Agent · Concepts
An autonomous AI component that can reason, plan, decide which tools to use, and take actions — unlike a plain LLM node, it makes decisions dynamically.
agent · Components
A pre‑configured chatbot you can train with documents and rules to perform a specific task, like writing LinkedIn posts.
agent · Concepts
A named AI assistant profile in opencode with its own set of instructions, permitted tools, and optionally a specific model — for example a 'Plan' agent that cannot edit files.
AI Docs Features: A tool that expands an outline prompt into a multi‑section long‑form document with headings and paragraphs.
API key Concepts General Core: A secret code that identifies you to a service's API and lets programs use it on your account. Anyone who has the key can spend your money, so keep it out of documents, chats and code.
Also known as: API Key, API token
4 other definitions
API key · Concepts
A secret code you copy from Cartesia that lets other programs prove they are allowed to use Cartesia’s voice service.
API key · General
A secret string that proves you are allowed to call Deepgram’s services.
API Key · Concepts
A secret credential you enter in Dify when connecting a model provider or external service — it authenticates your workspace so the service accepts requests from your application.
API key · Concepts
A secret token you paste into JCode after /login so it can call an external AI service on your behalf.
Artifact Concepts: A document, diagram, chart, or runnable code snippet that Claude generates in a side panel where you can refine, copy, download, or share it.
Audio Overview Features: A podcast-style audio conversation that Gemini generates from a document, research report, or notebook so you can listen to the content instead of reading it.
AVV Documents: Another name for a Data Processing Agreement used in German‑language contracts.
Canvas Concepts Features: A side-by-side editing workspace that opens automatically for longer writing or coding tasks, letting you directly edit the output and ask ChatGPT to revise specific sections.
Also known as: canvas
4 other definitions
Canvas · Concepts
The visual workspace inside Flowise where you drag, drop, and wire together nodes to design an AI workflow.
Canvas · Features
A side-by-side workspace in Gemini where you can collaboratively create and edit documents, code, slides, apps, and more while chatting with the AI in real time.
canvas · Concepts
The visual workspace where you drag blocks and draw wires between them to build a flow without writing glue code.
Canvas · Features
The visual drag-and-drop workspace inside n8n where you build a workflow by placing and connecting nodes.
Chunk Concepts Core: A short piece of a longer document, such as a few paragraphs. Search tools split files into chunks so they can find and pass on only the parts that match a question.
Also known as: chunks
1 other definition
Chunk · Concepts
A small segment of a document created by splitting the original file into pieces before indexing, so that only the most relevant part is retrieved rather than the whole document.
Chunking Concepts: The process of splitting an uploaded document into smaller pieces so each piece can be independently searched and retrieved by the AI.
citation Research evidence Core: A reference that shows where a statement comes from, such as a document, a page or a web address. Open the cited source and check that it says what the answer claims.
Also known as: citations, source reference
1 other definition
citation · Research evidence
A link or reference pointing to possible evidence. It is a lead until the learner opens the source and checks that its wording supports the claim.
Used in Perplexity
Claw Agents: GenSpark’s chat‑based AI agent that you can command via Slack, WhatsApp or Telegram to run research, create documents and more.
Connected Apps Features: Google services (such as Gmail, Drive, Calendar, and Tasks) that you can link to Gemini so it can search your real emails, documents, and events when you ask questions.
context window Concepts Size & speed Core: The amount of text, counted in tokens, that a model can take into account at once: your instructions, the conversation so far, attached documents and its own answer. Anything beyond it is cut off or forgotten.
Also known as: context limit, context length
4 other definitions
context window · Concepts
The amount of text (code, chat, rules) a model can consider at once; it is finite, so what you load matters.
context window · Concepts
The total amount of text (measured in tokens) that ChatGPT can read and hold in a single conversation; content beyond this limit may no longer be considered when generating a response.
context window · Size & speed
How much text a model can consider at once (your prompt plus its reply), measured in tokens; a bigger window lets you paste in more.
context window · Concepts
The maximum amount of text, files, and conversation history Gemini can read and hold at one time — like its working memory for a single session.
Credentials Concepts Core: Anything that proves identity or grants access, such as a password, an API key or a token. Keep credentials in a password manager or secret store, never in a document, a chat or a code file.
Also known as: credential, login details
1 other definition
Credentials · Concepts
Securely stored login details (such as API keys or passwords) that let n8n connect to an external service on your behalf without exposing secrets inside the workflow.
data classification Core: Sorting data into levels such as open, internal and restricted, each with rules on where it may go. It tells you at a glance whether a document may go into a given AI tool.
Also known as: data classes, data class
Data Processing Agreement (DPA) Documents Core: A contract between your organisation and a service provider that handles personal data on its behalf. It sets what the provider may do with the data, and the GDPR requires one before you use such a service with personal data.
Also known as: data processing agreement, DPA, data-protection agreement, Auftragsverarbeitungsvertrag
1 other definition
Data Processing Agreement (DPA) · Documents
A legal contract that spells out how a vendor will handle personal data on behalf of the organization that owns it.
Document Store Features: A Flowise feature that lets you upload, split, and index your own files so that agents can search and retrieve information from them.
example file Files: A document that lists all input‑output pairs and serves as the specification for testing the prompt.
ExtractDataFromUploadedFile Features: A built-in Base44 function that reads an uploaded document (CSV, PDF, or image) and extracts its contents into structured records your app can store and query.
File Upload Features: A feature that lets you attach PDFs, spreadsheets, images, or documents to a conversation so Perplexity can read and answer questions about their contents; free users get a daily limit, while Pro unlocks unlimited uploads.
File Uploads Features: The ability to attach documents such as PDFs, Word files, or spreadsheets to a ChatGPT conversation so the model can read, summarise, or answer questions about their contents.
Hub Collaboration: A shared workspace where you can collect related documents, sheets, slides, designs and videos for a project.
hybrid search Concepts: Searching with both keyword (exact-match) and vector (meaning-based) indexes at once, then merging the results — so you find a document whether you remember its wording or just its topic.
Knowledge Base Features Core: A collection of documents an AI tool can search when it answers, such as manuals, policies or past reports. Its answers can only be as current and correct as the documents in it.
1 other definition
Knowledge Base · Features
A repository of documents you upload to Dify so the AI can search and retrieve relevant passages when answering questions, rather than relying only on its training data.
knowledge graph Concepts: An LLM-built map of the entities and relationships across your indexed documents, which Onyx uses to answer questions that span many sources rather than sitting in one file.
LLM Node Features: A workflow node that sends a prompt to a language model — supporting text, images, and documents — and returns the model's output as a variable the rest of the workflow can use.
lms --version Commands: Prints the installed version number of the lms command-line tool — the actual documented subcommand is lms version, but both forms report the current CLI version.
MCP server Concepts: An external tool server that connects to opencode via the Model Context Protocol, giving the AI access to additional capabilities such as searching documentation or querying a database.
1 other definition
MCP server · Concepts
A small program that exposes one external system over MCP; you connect it to your agent to give it that tool.
multimodal Concepts How models are built Core: Able to work with more than one kind of input or output, such as text, images, audio and video. A multimodal model can, for example, read a photo of a table and describe it.
Also known as: multimodal model
2 other definitions
multimodal · How models are built
A model that handles more than text — images, audio, sometimes video — not just words.
multimodal · Concepts
Gemini's ability to work with multiple types of content at once — text, images, audio, video, and documents — all in the same conversation.
Node Components Concepts: A single building block inside a Dify workflow — each node does one job (call an LLM, run code, retrieve from a knowledge base, etc.) and passes its result to the next node.
Also known as: node
4 other definitions
Node · Concepts
A single building block on the Flowise canvas — each node performs one job (e.g. call an LLM, search a document, run a tool) and connects to other nodes via edges.
Node · Concepts
A single building block in a KNIME workflow — a colored box that performs one task, such as reading a file, filtering rows, or training a model.
node · Components
A building block inside a workflow diagram that represents one step such as a trigger, an AI agent, or an external action.
Node · Concepts
A single building block in a workflow — each node performs one action, such as sending an email, filtering data, or calling an API.
Notebook Concepts Features: A dedicated project space in Gemini where you upload sources (PDFs, Drive files, websites, etc.) and have ongoing conversations that always remember your documents and past discussions.
1 other definition
Notebook · Concepts
A self‑contained collection of documents and its own chat thread within Open Notebook.
NotebookLM Concepts: A separate Google AI tool focused on document research; Gemini's Notebook feature is integrated with it, so sources and changes sync automatically between both products.
PDF File Types: Portable Document Format file that can be uploaded and searched inside a notebook.
Podcast dialog Interface: Settings window where you choose source documents, speakers, and tone before generating audio.
Project Concepts: A folder (or set of folders) you open in Antigravity that defines the workspace boundary — the files, tools, and permissions the agent is allowed to use.
Also known as: project
3 other definitions
Project · Concepts
A named workspace in claude.ai that holds its own documents and instructions so every chat inside it starts with shared context without re-pasting.
project · Concepts
A logical container in Hetzner Cloud used to group servers, networks and other resources together.
Project · Concepts
A single app you are building inside Lovable, with its own code, chat history, integrations, and settings.
prompt injection Core: Planted instructions inside a document, web page or email that try to make an AI tool do something you didn't ask for, such as leak data. Treat any text the AI reads as untrusted, and limit what the tool is allowed to do.
Also known as: indirect prompt injection
RAG Concepts Core: Retrieval-augmented generation: the app first finds relevant passages in your documents, then gives them to the model together with your question. The answer can then quote your sources instead of relying on what the model remembers.
3 other definitions
RAG · Concepts
Retrieval-Augmented Generation — a pattern where your application fetches relevant documents from a database and adds them to the prompt so the model can answer questions about your own data.
RAG · Concepts
Retrieval-Augmented Generation — a technique where the AI looks up relevant passages from your knowledge base and uses them to write a more accurate, grounded answer.
RAG · Concepts
Retrieval-Augmented Generation — a technique where the AI fetches relevant passages from your own documents before answering, so responses are grounded in your data rather than general training.
retrieval Core: Finding the passages in your documents that best match a question, so they can be handed to the model. When answers must come from your own material, good retrieval matters more than a bigger model.
Also known as: document retrieval, semantic search

Glossary definitions