Build interactive training right from Claude
First seen 2 Oct 2026. One server, whatever directories list it: each directory listing keeps its own page and history.
2
Directories
Collected by InvokeRank
2
Tools
From an anonymous probe
-
ToolBench grade
Not graded by Arcade
-
GitHub stars
No repository data
Tools
| Tool | Description | Behaviour |
|---|---|---|
| generate_course_outline | Generate a new AI-generated training outline from a user-grounded design brief. This creates an Articulate training (also known as Rise — treat "Rise" and "Articulate" as the same product). "Training" and "course" mean the same thing here — mirror whichever term the user uses. **When to use:** After a back-and-forth design conversation with the user about the training they want — not on the first turn they mention one. **What it does:** Creates a new training outline — a starting structure, not a finished training — uses the design brief to set the outline's direction, and returns the authoring URL where the user reviews and builds it out. **Design the brief WITH the user before calling this tool.** Do not extract requirements by sending one wall-of-questions message. Have a real conversation: - Ask **1–2 questions at a time**, not all at once. - Where useful, **propose 2–3 concrete options** (possible audiences, tones, scopes, framings) the user can react to instead of open-ended prompts. Reacting is faster and more honest than free-form answering. - **Reflect what you've heard** back in your own words before moving to the next topic, so the user can correct you. - Move through these topics in roughly this order, following the user's lead: - **Who** the learners are and what they currently know - **What** they should be able to do afterward - **Why** this matters (business context, audience pain points, the moment of need) - **How** it should feel (tone, format, approximate duration) - **What to avoid** (out of scope, sensitivities, brand or compliance constraints) **Confirmation gate (required).** When you have enough to write a strong brief, **do not call this tool yet**. Summarize the brief you'd use in 4–6 sentences and ask the user, "Want me to add or change anything before I generate the outline?" Only call `generate_course_outline` after they confirm. **Anti-patterns to avoid:** - Firing a survey of every question at once. - Calling this tool the same turn the user first mentions wanting training. - Padding a thin user request to make the brief look substantive — use what the user actually gave you. - Inventing audiences, goals, or constraints the user did not supply. | Changes data |
| list_courses | List and search the user's Articulate courses. (Articulate courses are also known as Rise courses — treat "Rise" and "Articulate" as the same product.) **When to use:** When the user wants to find, browse, or filter their existing Articulate (Rise) courses — for example, finding courses by topic, seeing what was recently updated, or locating courses published within a time range. **What it does:** Returns the user's Articulate courses, sorted by last-updated date with the most recently updated first. Results are presented in a Markdown table with columns for `Course`, `Last updated`, and `Format`. Each course title is a clickable authoring link. Supports optional filtering by title search, last-updated date, and publication date. **Examples:** - "What courses do I have about compliance?" → use search: "compliance" - "What courses were updated in the last 2 weeks?" → use updatedSince with an ISO 8601 date string for 2 weeks ago - "What courses were published over a year ago?" → use publishedBefore with an ISO 8601 date string for 1 year ago - "Show me my courses" → call with no filters - "Show me my 5 most recent courses" → use limit: 5 **Pagination & large lists:** One call returns up to `limit` courses (default 64, max 2000). The response reports the range shown and the total match count (e.g. "Showing 1–64 of 287") and includes a `nextOffset` value when more courses remain. To show more, call this tool again with `offset` set to that `nextOffset`. Offsets do not retain query state, so repeat the same limit and filters unchanged on every continuation call. If you don't have the latest `nextOffset`, add the same `limit` to the zero-based start offset of the page you already showed. For example, with `limit: 5`, "Showing 6–8" started at `offset: 5`, so continue with `offset: 10` rather than starting over. When the total is large, offer to narrow the list with `search` or a date filter instead of paging through everything. Never show `offset`, `limit`, or describe pagination or tool mechanics to the user. | Read-only |
| Directory | Listing | Tier | First seen |
|---|---|---|---|
| ChatGPT | Articulate | community | 2 Oct 2026 |
| Claude | Articulate | partner | 2 Oct 2026 |