Skip to main content
Glama

Server Details

Create, edit and publish Educaplay educational games and quizzes from any AI assistant.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP · MCP 2025-11-25
URL
ChatGPT
Educaplay
Claude
–

TDQS

A3.6/5.0

Scored across 41 tools

Disambiguation4/5

Descriptions go to unusual lengths to draw boundaries between the near-duplicate pairs: illustrate_items vs set_item_images vs search_image_library, set_item_explanations vs change_item_explanations, classify_activity vs set_activity_audience vs add_label/remove_label. Most tools have a genuinely distinct purpose, but the sheer number of fine-grained media/explanation variants means an agent could still misselect in a few spots.

Naming Consistency4/5

Almost everything follows a verb_noun snake_case pattern (create_activity, get_activity_content, set_folder_guest_access, list_my_guests), which is highly predictable. Minor deviations like change_item_explanations vs set_item_explanations use different verbs for related operations, but the distinction is intentional and readable.

Tool Count2/5

41 tools is well past the heavy threshold and reflects an API surface split into many narrow operations (separate tools for explanations writing vs explanation display, image search vs image apply vs item images vs puzzle image). While each maps to real editor functionality, the set is too large to be well-scoped for an agent to navigate efficiently.

Completeness4/5

The surface covers the full activity lifecycle (create, read, content, settings, publish/unpublish, blockers), lists/folders CRUD, guest access, media, explanations, and Classroom export. Notable gaps remain: there is no delete_activity, and guest management is read-only (no invite/remove guest).

Publisher details

Operator
Create Play & Learn SL · Publisher source
Vendor relationship
First-party
Privacy policy
Not provided
Trust center
Not available
Restrictions
Requires an Educaplay account on the Academic or Commercial plan; on other plans the server answers 403 (license_required). No admin approval, regional limit or custom OAuth app: users sign in through OAuth 2.1 with dynamic client registration or a client ID metadata document.

Available Tools

41 tools
add_labelAdd labelB
Idempotent
Inspect

Add a label (tag) to an activity.

Inputs: label (required), activity_id (required).

attach_videoAttach YouTube videoAInspect

Add a YouTube video to the end of the timeline of a video quiz, playing whole. Only with a link the user gave, never an invented one. Returns the title, description and length of the video and the second where it starts and ends in the game, which is what the "at" of each question refers to.

Inputs: url (required), activity_id (required).

change_item_explanationsRemove or show item explanationsA
DestructiveIdempotent
Inspect

Remove explanations, or choose when the player sees them, without touching anything else. Call it ONLY when the user asked for it: "remove": the explanation goes away with its image or audio and cannot be undone; "remove_text": only its text goes, its image or audio stays (to take away only the image or audio, use remove_item_media); "show_always": shown after every answer; "show_on_mistake": shown only after a wrong answer. The result lists the items that changed and in "skipped" the ones that could not, with the reason.

Inputs: items (required), scope (required), action (required), activity_id (required).

classify_activityClassify activityA
Idempotent
Inspect

Fill in the classification of an activity that is still empty: its educational level, its knowledge area and its labels. It is the step that set_activity_content asks for when its result lists "level", "area" or "labels" in next_step.fields; the values come from next_step.catalog and from what the activity is about. It only fills what is empty and never changes a value already set: when the user asks to change the level, the area or the labels, use set_activity_audience, add_label or remove_label.

Inputs: activity_id (required), area, level, labels.

create_activityCreate activityAInspect

Create a new empty activity for the current user. Its result brings "content_schema": the schema, the writing guide and the limits of the content of that type, which is what set_activity_content needs next. Keep working on the activity already open: create a second one only when the user explicitly asks for another activity, never as a way around the type that is open.

Inputs: type (required), title (required), description (required), lang.

create_folderCreate listAInspect

Create a list for the current user (the Lists of My games). Lists cannot go inside other lists. On plans below Academic an account can have at most 3 lists, and creating one more fails with "folder_limit_reached". Same as POST users/{idUser}/folders/ with action=create of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response". The new list comes in "folder".

Inputs: name (required).

create_google_classroom_assignmentCreate Google Classroom assignmentAInspect

Create an assignment in Google Classroom with an Educaplay activity, in one of the Classroom courses of the user. It does exactly what sharing the activity to Google Classroom from the Educaplay website does: Educaplay creates the assignment, so the grade of each student reaches Classroom when they finish the game. As on the website, the activity can be one of the user's or one by another author. Call it only when the user asks to send or assign an activity to Google Classroom. Without course_id it creates nothing and returns the courses of the Google account connected to Educaplay: ask which one when it is not clear, then call again with its course_id. Each assignment counts against the Google Classroom limit of the plan of the user, the same as on the website.

Inputs: activity_id (required), draft, grade, course_id.

delete_foldersDelete listsA
DestructiveIdempotent
Inspect

Delete one or several lists of the current user (the Lists of My games). The games inside are not deleted: they stay in My games and in their other lists. Ids of lists of other users are ignored. Same as POST users/{idUser}/folders/ with action=delete of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response".

Inputs: ids (required).

describe_imageDescribe imageA
Read-only
Inspect

Describe one image of the media library. Pass the media_id returned by list_my_media or search_image_library. Use it only when the user explicitly asks about an image.

Inputs: media_id (required).

generate_field_audioGenerate spoken audioA
Destructive
Inspect

Generate the spoken audio of a part of the activity and save it into its texts. Call it ONLY when the user asked for audio in their own words: generating spends the text to speech quota of the user, so never add audio on your own initiative, not even to finish an activity you have just written. Say which part they named and nothing more: "elements" are the kinds of text they named, from "media_targets.audio" of the activity state ("the questions" is question, never question and choice; "the answers" is choice; "everything" is every element listed there), and "answers" narrows the choices to the correct or the incorrect ones. With "selection": "all_of_elements" every text of those elements gets its audio and you send no texts; when they named items by their place (question 2, the odd ones, the first five), add those positions in "positions". Only for a rule about the texts themselves (the capitals, the longest ones, one text they quoted), use "listed_texts" and send exactly those texts in "items". A text that already has audio keeps it unless existing is "replace"; one with a video or a youtube link is always left untouched. The audio speaks with the voice and the speed the user chose last in that language: send "audio_config" only when the user asked in this request for a voice or a speed, never on your own initiative (list_voices gives the voices, also to pick one by gender or accent). The result lists in "generated" every text that got its audio, with the element and the item it belongs to; when it has "voice" or "speed", tell the user which ones spoke.

Inputs: answers (required), elements (required), existing (required), language (required), +5 more.

get_activityGet activityB
Read-onlyIdempotent
Inspect

Read the current state of one activity: metadata, labels, settings, quality suggestions and publish blockers.

Inputs: activity_id (required).

get_activity_contentGet activity contentB
Read-onlyIdempotent
Inspect

Read the content items of one activity, paginated and in a readable form.

Inputs: activity_id (required), page, page_size.

get_activity_schemaGet activity content schemaA
Read-onlyIdempotent
Inspect

Get the JSON schema of the "content" argument of set_activity_content for one activity type, plus its writing guide and limits.

Inputs: type (required), include_settings.

get_folder_guest_accessGet guest access to listsA
Read-onlyIdempotent
Inspect

Tell who can open each of the given lists of the current user (the Lists of My games): "all" true means every guest of the user, "groupIds" and "guestIds" are the groups and the guests with access, and "guests" gives the names of those guests. Same as GET users/{idUser}/folderguests/?ids= of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response".

Inputs: ids (required).

get_publish_blockersCheck publishing blockersA
Read-onlyIdempotent
Inspect

Check whether an activity can be published and what is blocking it.

Inputs: activity_id (required).

illustrate_itemsIllustrate itemsA
Destructive
Inspect

Add images to the activity: for every text you send, it searches the Educaplay image bank and a stock photo library for what you describe, chooses the best match and puts it on that text, without touching the text or its audio. It is the way to add images, and you must call it ONLY when the user asked for images in their own words: never illustrate an activity on your own initiative, not even one you have just written. The user decides WHICH texts get an image and you send exactly those, every one of them: all the questions, every choice, only the correct choices, only the incorrect ones, the odd or the even items, the first five, one named text... whatever they said. Never reduce their selection, never pick a subset of your own and never ask which ones to start with: send the whole selection in this call, up to 40 entries, and call again in the same turn with the rest when there are more. Which texts can hold an image depends on the type: "media_targets.image" in the activity state lists them (for example question and choice, or word, hint, definition, card). A text that already has an image keeps it unless existing is "replace"; one with a video or a youtube link is always left untouched. The result lists in "applied" every text that got its image, with the element and the item it went to and, when known, what the image shows, and in "pending" the texts that are still waiting for theirs.

Inputs: items (required), answers (required), elements (required), existing (required), +1 more.

list_activity_typesList activity typesA
Read-onlyIdempotent
Inspect

List the activity types the assistant can create and edit, with their localized name, what each one is best for, payload key and capabilities. The order changes on every call and "how_to_choose" says how to pick the type of a new activity from it. Not needed when the user already named the type: create_activity takes it directly.

Inputs: only_ai_capable.

list_folder_activitiesList the games of a listA
Read-onlyIdempotent
Inspect

List the games inside one list of the current user (the Lists of My games), in the order My games shows them. A list only shows its published games. Same as GET users/{idUser}/activities/?folder= of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response". While "hasMore" is true, pass "nextCursor" as "before" to get the next page.

Inputs: folder (required), q2, sort, type, +2 more.

list_my_activitiesList my activitiesB
Read-onlyIdempotent
Inspect

List the activities of the current user, newest first, with pagination.

Inputs: page, type, query, page_size, +1 more.

list_my_foldersList my listsB
Read-onlyIdempotent
Inspect

List the lists of the current user (the Lists of My games), newest first. Each list brings its number of games ("numResources"), "shareUrl", the public link to share the list, "url", where the user opens it in My games, and "guestAccess", the guests and groups that can open it. Lists cannot contain other lists. Same as GET users/{idUser}/folders/ of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response". While "hasMore" is true, pass "nextCursor" as "before" to get the next page.

Inputs: before, length.

list_my_guest_groupsList my guest groupsA
Read-onlyIdempotent
Inspect

List the groups of guests of the current user, with how many guests each one has and the games and lists the group can open. Use the ids with set_folder_guest_access. Same as GET users/{idUser}/guests/groups/ of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response".

Inputs: no named parameters.

list_my_guestsList my guestsA
Read-onlyIdempotent
Inspect

List the guests of the current user: the people they invite to play with an access code, without an Educaplay account. Each guest brings its id, name, access code and link, its groups, and the games and lists it can open. Use the ids with set_folder_guest_access. Same as GET users/{idUser}/guests/ of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response". While "hasMore" is true, pass "nextCursor" as "before" to get the next page.

Inputs: q2, group, before, knownOnly, +1 more.

list_my_mediaList my mediaA
Read-onlyIdempotent
Inspect

List the media files the user already has in their activities, newest first and with no relation to what they show, so that the user can point at one of them. To choose an image for a text use search_image_library instead: this list cannot tell you what each file contains.

Inputs: kind, page, query.

list_voicesList voicesA
Read-onlyIdempotent
Inspect

List the voices that can speak the audio of a language: the same ones the audio window of the editor offers, each with its name, gender and locale (es-ES Spain, es-MX Mexico, es-US United States...), and "high_quality" for the ones the editor marks so. It also gives the speeds with the names the editor shows and, in "user_preference", the voice and speed the user chose last in that language, which generate_field_audio uses when it gets no audio_config. Call it when the user asks which voices there are, or asks for a voice by its gender or accent, to pick its name. With "fallback_language", the language has no voices of its own and its audio is read by these.

Inputs: language (required).

publish_activityPublish activityA
Destructive
Inspect

Publish an activity so that it can be played and shared. It checks the publishing blockers itself and, when the activity is not ready, fails with their list.

Inputs: activity_id (required).

remove_item_mediaRemove item mediaA
Destructive
Inspect

Take the image or the audio away from texts of the activity, leaving the text and everything else as it is. It cannot be undone. Call it ONLY when the user asked to remove, delete or take away the image or the audio of texts they name, or to swap one medium for the other ("change the image of Paris for an audio": first add the audio, then remove the image). Never use it to change an image for another image or an audio for another audio: that is "existing": "replace" in illustrate_items, set_item_images or generate_field_audio. Send exactly the texts the user named, up to 40 per call. The result lists in "removed" every text that lost it and in "skipped" the ones that had none.

Inputs: kind (required), items (required), activity_id (required).

remove_labelRemove labelB
DestructiveIdempotent
Inspect

Remove a label (tag) from an activity by its name.

Inputs: label (required), activity_id (required).

rename_folderRename listA
DestructiveIdempotent
Inspect

Rename a list of the current user (the Lists of My games). Same as POST users/{idUser}/folders/{idFolder}/ with action=rename of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response".

Inputs: name (required), folder (required).

search_image_librarySearch image libraryA
Read-onlyIdempotent
Inspect

List image candidates WITHOUT adding anything to the activity. Use it only when the user asks to see or to choose among options ("show me options for...", "which images are there of..."): to put images on the activity use illustrate_items, which finds and applies them in one step. Send one query per concept, up to 10 in the same call: each one is answered with candidates from the Educaplay image bank and from a stock photo library, every item with its "source" and a "title" that says what it shows. A call with a single query answers with more candidates and adds the images the user already has. Returned image ids are used in set_item_images.

Inputs: queries (required), page.

set_activity_audienceSet activity audienceA
DestructiveIdempotent
Inspect

Set who the activity is aimed at: content language, target audience and knowledge area. Call it only when the user asks for one of these; a title or a description never needs it.

Inputs: activity_id (required), area, lang, audience.

set_activity_contentSet activity contentA
Destructive
Inspect

Add or replace the content of an activity. The "content" object must follow the schema of the activity type: the "content_schema" that create_activity returned, or get_activity_schema for an activity that already existed. Never guess it.

Inputs: content (required), activity_id (required), mode, audio_config.

set_activity_foldersPut games in listsA
DestructiveIdempotent
Inspect

Put games into lists of the current user (the Lists of My games), take them out or move them from one list to another. With mode "add" the games go into the lists in "folders" and stay in the ones they were in; with "remove" they leave those lists; with "set", the default, each game ends up in exactly the lists in "folders", so "set" with an empty "folders" takes it out of all its lists. To move a game, send "set" with its new lists, or "remove" from one list and then "add" to the other. Taking a game out of a list never deletes it. Lists only show published games: with scope "all" only published games are taken, while games chosen by id go in whatever their state, and the unpublished ones do not show in the list. Ids of lists of other users are ignored. Choose the games by id, at most 50 per call, or with scope "all", every game that matches the filters ("folder", "type", "q2", "state") except those in "exclude": then each call handles one batch, and while "hasMore" is true you call again with "nextCursor" as "before". Same as POST users/{idUser}/activities/ with action=setFolders of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response". There "ok" lists the games changed and "failed" the ones that could not be, with the reason.

Inputs: q2, ids, mode, type, +7 more.

set_activity_privacySet activity privacyA
Destructive
Inspect

Change who can reach the activity: its visibility and whether the Top 10 ranking of scores is hidden. Call it only when the user asks to change one of the two.

Inputs: activity_id (required), privacy, hide_results.

set_folder_guest_accessSet guest access to listsA
DestructiveIdempotent
Inspect

Give guests access to lists of the current user (the Lists of My games), or take it away. Guests are the people the user invites to play with an access code; access goes to one guest, to a group of guests or to all of them. Get the ids of guests and groups from list_my_guests and list_my_guest_groups. Lists of other users come back in "failed" as "forbidden", and only the first 50 list ids are used. Same as POST users/{idUser}/folders/ with action=setGuestAccess of the Educaplay REST API, for the current user: its JSON body, or its error body, comes in "response".

Inputs: ids (required), add, remove.

set_item_explanationsWrite item explanationsA
DestructiveIdempotent
Inspect

Write or rewrite the explanation of items that already exist, leaving everything else as it is: their texts, their images and audio, and the image or audio of that explanation. Call it ONLY when the user asked to add, write or change explanations or feedback of items they name, or of all of them; never set_activity_content "replace" for that. Asking when explanations are shown ("set the explanation of question 3 to always show", "only on mistake"), or to remove some, is not asking to write them: that is change_item_explanations alone, never rewrite an existing explanation then, and never fill the items that have none, offer it instead. Up to 40 items per call. The result lists in "written" every item that got it ("replaced" when it already had one) and in "skipped" the ones that could not, with the reason.

Inputs: items (required), activity_id (required).

set_item_imagesSet item imagesA
Destructive
Inspect

Put one specific image, chosen by its id, on texts of the activity, without touching the text or its audio. Only for an image the user pointed at: one of their own files (list_my_media) or a candidate they chose from search_image_library. When the user simply wants images on their texts, use illustrate_items, which finds them by itself. Call it ONLY when the user asked for images in their own words. Which texts can hold an image depends on the type: "media_targets.image" in the activity state lists them (for example question and choice, or word, hint, definition, card). Send the text of that exact element copied from get_activity_content: to illustrate a choice send the choice, to illustrate a question send the question. A text that already has an image keeps it unless existing is "replace"; one with a video or a youtube link is always left untouched. The result says, for every image, which element and which item it went to.

Inputs: images (required), existing (required), activity_id (required).

set_item_text_visibilityShow or hide item textsA
DestructiveIdempotent
Inspect

Show texts of the activity as their image alone, or show their text again. Call it ONLY when the user asked for it in their own words: "leave only the flag", "remove the word and keep the picture", "only images in the left column" (hidden), or "put the names back" (shown). Hiding keeps the text stored, so the item can still be named and shown again, and it never touches the image or the audio. Only a text that already has an image can be hidden: when the user wants only images on texts that have none, first add them with illustrate_items and then call this tool in the same turn. Which texts can show only their image depends on the type: "media_targets.image_only" in the activity state lists them; an empty list means the type always shows its texts. Send exactly the texts the user named, up to 40 per call. The result lists in "hidden" or "shown" every text that changed and in "skipped" the ones that could not, with the reason.

Inputs: items (required), visibility (required), activity_id (required).

set_puzzle_imageSet puzzle imageA
Destructive
Inspect

Put the picture of a puzzle: the image the player puts together, cut into pieces whose number follows its shape and the piece level. Only for a puzzle (its state has "puzzle"); images on its questions and answers go with illustrate_items. Source "attachment": an image the user attached to the conversation for the puzzle, by its attachment_id. Source "search": an image that shows what you describe in "shows", found in the Educaplay image bank and a stock photo library. Source "image_id": the one the user chose from search_image_library or list_my_media. A puzzle that already has its picture keeps it unless existing is "replace", and with "search" a replacement is always a different picture. The result says what the picture shows, when known, and how many pieces the puzzle has.

Inputs: source (required), existing (required), activity_id (required), shows, +2 more.

unpublish_activityUnpublish activityA
Destructive
Inspect

Unpublish an activity so that it can no longer be played. It stays editable and can be published again.

Inputs: activity_id (required).

update_activityUpdate title and descriptionA
DestructiveIdempotent
Inspect

Update the title and the description of an activity. Send only the one you want to change. An activity needs both to be published, and set_activity_content reports in next_step when they are still empty. The audience and the visibility have their own tools.

Inputs: intent (required), activity_id (required), name, description.

update_activity_settingsUpdate game settingsA
DestructiveIdempotent
Inspect

Update game settings of an activity (time limit, attempts, shuffling...). Get the editable fields with get_activity_schema(include_settings: true).

Inputs: settings (required), activity_id (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      C
      maintenance
      Enables an AI assistant to publish flashcard study sets directly into the Vocabit app and then read back detailed learner results, including weak and untouched cards, so future sets can be tailored to actual performance.
      8
      47 npm
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Connects AI assistants to a children's education platform so teachers, parents and content authors can create courses, lessons and interactive exercises, manage lesson calendars, review and grade homework, and track student progress through natural conversation. It also supports building educational games, media libraries, and age-targeted recommendation strategies.
      MIT
    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources