Glossary
This glossary defines the words SlideMind uses in the portal and in these guides. Each entry links to the page that explains the topic in full.
Activity type: what a slide asks the learner to do: Q&A, Chat or Capture. See Activity types.
AI mode: how a workspace reaches the AI, either on SlideMind Credits or on Your own API keys. You choose it in Settings › AI & models. See Choosing how AI is funded.
AI notice: an optional message the SlideMind widget shows before the first exchange, asking learners to acknowledge that the interaction sends content to an AI provider. See AI notice.
AI summary: a few lines the AI writes after each Chat reply, so a later slide can build on the conversation. It is one of the exchange memory options. See Saving values.
Allowed origins: the setting that holds your allowed websites. See Allowed websites.
Allowed websites: the sites allowed to use your published slides, listed in Settings › Integrations › Allowed origins. An empty list allows every site. See Allowed websites.
Archived: see Status.
Authoring tool: the software you build your courses with: Articulate Storyline, Rise 360, Adobe Captivate, Lectora, dominKnow | ONE or iSpring Suite. See Integration overview.
Authoring tool interface: the interface in which the learner uses your authoring tool’s own fields. The integration code reads and writes authoring tool variables. See Integration overview.
Authoring tool variable: a variable you create in your authoring tool, such as a Storyline text variable, that the integration code reads or writes. See Variables in prompts.
BYOK (“bring your own key”): running a workspace on your own provider keys, included from the Pro plan. The Activity screen uses this label for the cost of those calls. See Provider key.
Capture: the activity type that saves values from your authoring tool into learner memory, without calling the AI. See Activity types.
Chat: the activity type for a short conversation of several turns, with an optional turn limit. See Conversations (Chat).
Config UID: the public identifier of a slide, written in its integration code. It tells SlideMind which slide answers; when you delete the slide, its config UID stops responding. See Changing a slide after installation.
Course: the top level of organization in a workspace. A course carries a theme, objectives and content that the slides below it inherit. See Courses and modules.
Course-level slide: a slide attached to a course but to no module. It inherits only the course context. See Courses and modules.
Destination: where a slide runs, such as Articulate Storyline, a Storyline Web Object, a Rise 360 Code block or the HTML widget. It decides which integration code SlideMind generates. See Integration overview.
Draft: see Status.
Editor: a workspace role (see Owner, Editor, Viewer). The SlideMind screen where you configure a slide is the slide editor.
Embedded capture: a Capture slide delivered through the SlideMind widget. It shows nothing and saves values only when a script on the host page sends them. See HTML widget (iframe).
Exchange memory: an option of Q&A and Chat slides that keeps the learner’s answer, the feedback, the conversation or an AI summary in the identified learner’s memory after each reply. See Saving values.
Extra workspace: a workspace bought on top of those a Pro or Agency plan includes, added to the subscription with Change plan. See Plan and subscription.
Fair-use limit: the number of learner messages a workspace can receive each month, set by the plan. It protects against abuse and is never billed; SlideMind support raises it on request. See Plans and limits.
Fallback: the text used when a placeholder’s value is missing or empty, set under If empty, show on the placeholder’s pill and written after |. With {{memory.first_name|there}}, SlideMind uses “there” when the first name is unknown. An empty fallback removes the placeholder. See Variables in prompts.
Identified learner: a learner SlideMind recognizes because the course was launched through SCORM, xAPI or cmi5 in an LMS, or because the embed URL carries an sm_user parameter. Only identified learners have a learner memory. See Identifying learners.
Independent slide: a slide attached to no course and no module. It works on its own, and you can attach it later. See Courses and modules.
Inherited context: the course theme, objectives and content, and the module subject and context, that every slide below them inherits. A slide inserts them in its instructions with placeholders such as {{course_theme}}, and can override any of these values. See Courses and modules.
Instructions: the text that tells the AI its role and how to respond, also called the system prompt. Learners never see it. See Write the AI instructions.
Integration code: the code that the Installation step of a slide generates for you to add to your course. For a Storyline Web Object, it is a folder you download; for a Storyline Q&A or Chat, it can also be a ready-made slide (.story). See Integration overview.
Interface: how the learner meets a slide: through the authoring tool interface or the SlideMind widget. See Integration overview.
Learner data: the screen that lists the values stored for each identified learner. Workspace owners can open, correct, export or purge them. See Learner data.
Learner memory: the values SlideMind keeps for each identified learner, such as an answer or a captured value, so other slides can reuse them. See How learner memory works.
Max tokens: the maximum length of an AI reply, counted in tokens. See Models and response settings.
Memory Hub: the catalog of the workspace’s memory variables, with their scope, and the tools that generate code to save them from, or load them into, an authoring tool. See Memory Hub.
Memory key: the name under which a value is stored in learner memory, such as profession. Other slides read it back with {{memory.profession}}. See Saving values.
Module: a group of related slides that share a subject and supporting context. A module can belong to a course or stay independent. See Courses and modules.
Opening message: the first message learners see when a Chat opens. It sets the scene but doesn’t call the AI by itself; SlideMind sends it to the AI at the head of the conversation. See Conversations (Chat).
Owner, Editor, Viewer: the three roles in a workspace. Owners manage the workspace, including AI, keys, members and learner data; editors create and test content; viewers can read content but can’t create, change or test slides. See Members and roles.
Ownership transfer: handing a workspace over to another SlideMind account, with its content, learner memory, members and embed codes. The recipient accepts from an e-mail link; their plan and SlideMind Credits apply from then on. See Transfer ownership of a workspace.
Placeholder: a name between double braces, such as {{course_theme}}, that SlideMind replaces with a value when the slide runs. See Variables in prompts.
Pause: stopping a paid subscription for 1 to 6 months. Nothing is billed, the workspaces are read-only and slides tell learners the course is unavailable. See Plan and subscription.
Plan: what the workspace owner’s account can create and use: Discovery (free), or the Starter, Pro or Agency subscriptions. It applies to every workspace the owner owns. See Plans and limits and Plan and subscription.
Provider key: your own API key from an AI provider, stored encrypted in the workspace. Slides use it when the workspace runs on Your own API keys. See Your own provider keys.
Published: see Status.
Q&A: the activity type in which the learner answers one question and gets one AI answer. See Activity types.
Question: in a Q&A slide, the question shown to the learner and sent to the AI with each answer. With the authoring tool interface, keep it identical to the question on your screen. See The slide editor.
Reasoning effort: for models that support it, how much the model reasons before it answers. Higher effort generally increases latency and cost. With SlideMind Credits, Automatic means a low effort. See Models and response settings.
Reference document: a PDF, Word, PowerPoint, text or Markdown document whose text you add to the workspace library and link to a course, a module or a slide. The AI reads it when learners ask. SlideMind keeps only the text, never the file. See Reference documents.
Response example: an example answer that shows the AI the tone, structure and level of detail you expect. On a slide, it is the Example answer field; courses and modules have their own Response example field. See Write the AI instructions.
Return variable: a named, typed field (text, number, boolean or enum) that the AI fills in a structured result, such as score or passed. In an authoring tool, it arrives in a variable named after the answer variable, such as aiResp_s2_quiz_score. See Results: plain text or structured.
Sandbox: the workspace screen where you build and try a slide draft, from a blank slide or a copy of a saved slide, without saving anything. See Sandbox.
Scope: where a memory value can be read back: Global (the whole workspace), Course, Module or Slide. See Saving values.
Slide: one AI interaction configured in SlideMind, which you place on a screen of your course. It is not a slide of your authoring tool. See Create a slide.
Slide editor: the SlideMind screen where you configure a slide step by step, try it, and get its integration code. See The slide editor.
Slide variable: a value your authoring tool sends with the learner’s answer, such as a quiz score, which the instructions use as {{score}}. See Variables in prompts.
SlideMind Credits: prepaid usage that funds AI calls on SlideMind’s managed keys, with no API key to provide. Credits belong to the account of the workspace owner. See SlideMind Credits.
SlideMind widget: the answer box or chat window hosted by SlideMind and embedded in your course, through an iframe, a Rise 360 Code block or a Storyline Web Object. See What learners see.
sm_user: a parameter of the embed URL that carries a learner identifier, so a learner outside an LMS can be identified. See Identifying learners.
Status: the state of a slide: Draft, Published or Archived. Only a published slide answers your course. See The slide editor.
Structured result: an AI reply made of named fields in JSON, instead of free text. See Results: plain text or structured.
Temperature: how varied the AI’s replies are, from focused to creative. Some models and reasoning settings don’t offer it. See Models and response settings.
Template: a reusable starting point for a slide, which copies its instructions, activity type and return variables. Built-in templates are maintained by SlideMind; custom templates belong to your workspace. See Templates.
Token: the unit AI models use to measure text. Instructions, messages and replies are counted in tokens, which drive the cost of each call. See Models and response settings.
Try it: the panel of the slide editor that sends real test messages to the AI with the slide’s current settings. Nothing is saved. See Try it.
Turn: in a Chat, one learner message plus one AI reply. See Conversations (Chat).
Variable: a named value that a slide reads or returns. Depending on where the value comes from and where it goes, see Inherited context, Slide variable, Learner memory, Return variable and Authoring tool variable. See Variables in prompts.
Viewer: a workspace role. See Owner, Editor, Viewer.
Workspace: the space that holds the courses, provider keys, learners, logs and members of one team or client, separately from any other workspace. See Workspace settings.