Getting Started
Installation
Section titled “Installation”Requirements
Section titled “Requirements”| Requirement | Minimum |
|---|---|
| WordPress | 6.0+ |
| PHP | 8.0+ |
| PHP extension | OpenSSL (for API key encryption) |
| AI provider | An API key from OpenRouter, OpenAI, Anthropic, Mistral, or Google AI Studio |
Note: Most WordPress hosts already have OpenSSL enabled. SlideMind checks on activation and displays a notice if it’s missing.
Install the plugin
Section titled “Install the plugin”From a ZIP file:
- In your WordPress admin, go to Plugins > Add New > Upload Plugin.
- Upload the
slidemind.zipfile and click Install Now. - Click Activate.
Via FTP / File Manager:
- Extract the
slidemindfolder. - Upload it to
/wp-content/plugins/. - In the WordPress admin, go to Plugins and activate SlideMind.
After activation, a new SlideMind menu appears in the WordPress admin sidebar.
Get your API key
Section titled “Get your API key”SlideMind supports five LLM providers. OpenRouter is the easiest way to start — it’s a unified gateway with access to 200+ models through a single API key. You can also connect directly to OpenAI, Anthropic, Mistral, or Google (Gemini).
Quick start with OpenRouter
Section titled “Quick start with OpenRouter”- Go to openrouter.ai and sign up.
- Add credits to your account (pay-per-use — most models cost fractions of a cent per interaction).
- Go to Keys (or visit openrouter.ai/keys), click Create Key, and copy it.
Or use another provider
Section titled “Or use another provider”| Provider | Where to get a key |
|---|---|
| OpenAI | platform.openai.com/api-keys |
| Anthropic | console.anthropic.com/settings/keys |
| Mistral | console.mistral.ai/api-keys |
| Google (Gemini) | aistudio.google.com/apikey |
Enter the key in SlideMind
Section titled “Enter the key in SlideMind”- In WordPress, go to SlideMind > Settings.
- Select the provider tab and paste your API key.
- Click Test Connection to verify it works.
- Click Save Settings.
Your key is encrypted at rest using AES-256-CBC and never sent to the browser. All AI requests go through your WordPress server.
Recommended: For stronger encryption, add a custom encryption key to your
wp-config.php:define('SM_ENCRYPTION_KEY', 'your-random-64-character-string-here');If omitted, SlideMind auto-generates and stores an encryption key in the database — but a
wp-config.phpkey is better for production.
First slide in 5 minutes
Section titled “First slide in 5 minutes”The Wizard is the fastest way to create a configuration. It walks you through four steps.
Go to SlideMind > Add to launch the Wizard.
Step 1 — Hierarchy & Mode
Section titled “Step 1 — Hierarchy & Mode”Set up where your slide lives and how it behaves.
- Course (optional) — Select an existing course or create a new one. Courses are top-level containers that group related modules.
- Module (optional) — Select an existing module or create a new one. Modules group slides under a course.
- Slide title (required) — Give your slide a descriptive name (e.g. “Introduction Quiz Feedback”).
- Slide number (optional) — Match it to the slide number in your e-learning module.
- Widget mode — Choose one of three modes:
| Mode | Description | Best for |
|---|---|---|
| Q&A (default) | Single-turn: the learner asks one question and gets one answer. | Feedback, evaluations, quick interactions |
| Chat | Multi-turn conversation with an optional turn limit. | Tutoring, exploration, guided discussions |
| Capture | Saves e-learning variables to learner memory — no AI involved. | Collecting learner data, pre-populating context for later slides |
Note: Capture mode requires a Pro or Agency plan.
Step 2 — AI Model
Section titled “Step 2 — AI Model”Capture mode: This step is automatically skipped — Capture doesn’t use an AI model.
Configure which AI model to use and how it should behave.
- LLM Provider — Choose the API provider, or inherit from the course/global settings.
- Model — Pick from the dropdown (grouped by provider: Anthropic, OpenAI, Google, Meta, etc.). Each model shows its estimated cost per interaction.
- Temperature — Controls creativity (0.0 = deterministic, 2.0 = very creative). Default: 0.7.
- Max tokens — Maximum length of the AI response (1—128,000). Default: 1,024.
Tip: Settings are pre-filled with your defaults from SlideMind > Settings. You can override them per slide.
Step 3 — System Prompt / Variables
Section titled “Step 3 — System Prompt / Variables”What you see here depends on the mode you selected.
Q&A and Chat modes — System Prompt:
- Interaction type — Choose a template to start from (Socratic Tutor, Quiz Feedback, Roleplay, Open Question Feedback, Formative Evaluation, Guided Exploration) or write your own from scratch.
- Prompt editor — Write or edit the system prompt. Use
{{variables}}for dynamic content (e.g.{{learner_name}},{{question}}). - Variable bar — Click any variable to insert it into the prompt. Use
@in the editor for autocompletion. - Generate with AI — Let an AI draft your system prompt based on your pedagogical objectives.
Chat-specific options: turn limit, closing message, graceful close (the AI wraps up the conversation naturally on the last turn), cumulative cost projection.
Q&A-specific options: return variables — structured data (text, number, boolean, enum) sent back to your e-learning module for scoring or branching.
Capture mode — Variable Definitions:
In Capture mode, this step is titled Variables instead of “System Prompt”. No AI is involved — the slide saves e-learning editor variables into learner memory for reuse in other slides.
Define the variables to capture from the e-learning editor. For each variable, fill in three fields:
| Field | Description | Example |
|---|---|---|
| Memory key | The name SlideMind uses to store and retrieve the value. Lowercase letters, digits, and underscores. | profession |
| Scope | Where the variable is accessible: Global (everywhere), Course (this course only), or Module (this module only). Choose Global if in doubt. | Global |
| Editor variable | The name of the variable in your e-learning editor (Storyline, Captivate, etc.). | Learner_Profession |
Click + Add a memory variable to add more rows.
How it works:
Example: You want to record the learner’s profession so it can be reused in other courses. In your e-learning editor (Storyline, Captivate…), create a variable called
Learner_Profession. Then configure a capture here:
- Memory key:
profession— the name SlideMind uses to store and retrieve this value- Scope: Global — so it’s accessible across all courses
- Editor variable:
Learner_Profession— the variable name in your e-learning editorOnce published, when the learner reaches this slide, the value is saved automatically. Any other slide can then use
{{memory.profession}}in its prompt to personalize the AI response.
Step 4 — Review & Test
Section titled “Step 4 — Review & Test”Review everything and test it live before saving.
- Summary — The full configuration at a glance.
- Prompt preview (Q&A / Chat) — The rendered prompt with test variable values.
- Capture summary (Capture) — The variables that will be captured.
- Test (Q&A / Chat) — Send a test message to the AI and see the response, along with debug info (tokens used, cost, response time).
- Create — Click Create the configuration to save.
After creation, you land on the slide edit page where you can copy the integration code for your e-learning editor.
Understanding the hierarchy
Section titled “Understanding the hierarchy”SlideMind organizes content in three levels. This structure mirrors how e-learning content is typically organized and supports inherited context and learner memory scoping.
Course └── Module └── SlideCourse
Section titled “Course”The top-level container. A course represents an entire training program.
| Field | Purpose | Variable |
|---|---|---|
| Title | Name of the course | — |
| Theme | The overall subject | {{course_theme}} |
| Objectives | Learning objectives | {{course_objectives}} |
| Reference content | Documentation the AI can use to answer questions | {{course_content}} |
| Response example | Define the desired response format and tone for the whole course | {{course_response_example}} |
| LLM Provider | Override the global provider for this course | — |
| API key | Optional dedicated key for cost tracking per course | — |
| Status | Published or Draft | — |
Example: “Pediatric Social Care — Level 1”
Module
Section titled “Module”A group of related slides within a course.
| Field | Purpose | Variable |
|---|---|---|
| Title | Name of the module | — |
| Parent course | Which course this module belongs to | — |
| Theme | The module’s specific topic | {{module_theme}} |
| Objectives | Module-level learning objectives | {{module_objectives}} |
| Reference content | Supporting content or context | {{module_content}} |
| Response example | Define the response format specific to this module | {{module_response_example}} |
| API key | Optional dedicated key (overrides course key) | — |
| Status | Published or Draft | — |
Example: “Module 3 — Communication Techniques”
The individual interaction point — this maps to a specific slide or screen in your e-learning module.
| Field | Purpose | Modes |
|---|---|---|
| Hierarchy | Parent course and module (both optional) | All |
| Title | Name of the slide | All |
| Slide number | Match to the e-learning slide number | All |
| Widget mode | Q&A, Chat, or Capture | All |
| LLM Provider | Override the provider (inherits from course or global) | Q&A, Chat |
| Model | Which AI model to use | Q&A, Chat |
| Temperature | Creativity slider (0.0—2.0) | Q&A, Chat |
| Max tokens | Maximum response length | Q&A, Chat |
| System prompt | Instructions for the AI, with template selection | Q&A, Chat |
| Response example | Expected format and tone for this specific slide | Q&A, Chat |
| Turn limit | Max exchanges before closing (0 = unlimited) | Chat |
| Closing message | Displayed when the turn limit is reached | Chat |
| Graceful close | AI naturally wraps up on the last turn | Chat |
| Return variables | Structured data (text/number/boolean/enum) sent back to the editor | Q&A |
| Persistent memory | Capture editor variables to learner memory | All |
| Auto-save exchange | Save learner input and/or AI response to memory | Q&A, Chat |
| Internal notes | Admin-only notes (not sent to the AI) | All |
| Status | Published or Draft | All |
Example: “Slide 12 — Active Listening Roleplay”
Inheritance cascade
Section titled “Inheritance cascade”Context flows downward automatically. When a learner interacts with a slide, SlideMind builds the full prompt by combining:
- Course context (theme, objectives, content, response example)
- Module context (theme, objectives, content, response example)
- Slide prompt and settings
You write the pedagogical context once at the course or module level, and every slide below it inherits that context. No need to repeat yourself.
The API key also cascades: Slide > Module > Course > Global settings.
The LLM provider cascades similarly: Slide > Course > Global settings.
Do I need all three levels?
Section titled “Do I need all three levels?”No. Both course and module are optional:
- Slide only — Works fine for standalone interactions or quick tests.
- Course + Slide — Good for a simple training with no module structure.
- Course + Module + Slide — Full hierarchy for complex programs with learner memory scoping.
Next steps
Section titled “Next steps”- Authoring Tool Guides — Integration code for Storyline, Captivate, Lectora, DominKnow, and HTML Widget
- Core Features — Learner Integration, Sandbox, Templates, Code Generator, Wizard
- Learner Memory — Persist data across slides, modules, and courses
- Variables — Variable system, profile variables, return variables
- Import / Export — Move configurations between WordPress sites
- Administration — Dashboard, analytics, settings, CORS
- Security & Privacy — Proxy architecture, encryption, GDPR