create_deck
ChatGPTCreate an EMPTY Kunjani deck (deck.name required). Typical flow: create the deck, seed its outcomes with create_outcome, then add activities with publish_activity_batch. publish_activity_batch can also create the deck and its activities in one call, so use this only when you want an empty deck first. Share deck.teams_url from the response with the user when they are in Microsoft Teams or M365 Copilot, otherwise deck.url.
create_outcome
ChatGPTAdd one learning outcome to a Kunjani deck. Write outcomes as specific, observable, learner-facing statements of what the player should be able to do after the deck (for example "Recognise the reputational impact of poor service"), not broad topics. A deck wants 3 to 6 of them, and every activity should advance at least one. outcome.description is required.
create_question
ChatGPTAdd one activity to a Kunjani deck. Use publish_activity_batch for bulk authoring and this for a single addition. Follow the three-block rule: text is the player-facing trigger and must never reveal the answer, answer is the AI grader's answer key, and assessment_notes are flexible grading rules (accept synonyms and real-world examples, state the minimum valid points, no rigid tier rubrics) plus any media transcript, because the grader is text-only. question.suit_name, question.text and question.answer are required, and suit_name must name a suit on the deck (Jolt, Advance, Mystery, Oops, Explain, Demonstrate). Media is URL-based: picture_url and picture_2_url are the two player-facing slots, and facilitator_picture_url is the response slot. Each accepts a public image URL (max 5 MB), or an audio or mp4 video URL (max 16 MB). Video reaches the player but is never shown to the AI assessor. OMIT question.name so Kunjani auto-numbers within the suit (J1, O1, D1); only set a name to pin a slot.
delete_deck
ChatGPTPERMANENTLY delete one Kunjani deck after the client has obtained the user confirmation required for this destructive action. The deck and everything on it are deleted in this call.
delete_outcome
ChatGPTPERMANENTLY delete one learning outcome from a Kunjani deck and detach it from every activity on that deck. The activities themselves are not deleted. Only outcomes the deck owns can be deleted.
delete_question
ChatGPTPERMANENTLY delete one activity. It is removed from EVERY deck it is shared with, not just this one. The client must obtain the user confirmation required for this destructive action before invoking the tool. The activity is deleted in this call.
get_deck
ChatGPTGet one Kunjani deck by id, including its suits and question_order, the display order the deck overview and games use (which can differ from creation order). Read-only.
get_question
ChatGPTGet one activity on a Kunjani deck by id. Read-only.
list_decks
ChatGPTList Kunjani decks (id, name, description, dice option, suits, question order). Use it to find an existing deck and reuse its id with publish_activity_batch or the manage_* tools for an idempotent update instead of creating a duplicate. scope "mine" (default) returns the decks the user owns or has been given access to; "public" returns the curated public library, which is a separate browse surface and is NOT part of anyone's own deck list; "all" returns both. Read-only.
list_outcomes
ChatGPTList the learning outcomes on a Kunjani deck and the activities linked to each. Outcomes listed with kind "via_questions" belong to another deck and cannot be changed from here. Read-only.
list_questions
ChatGPTList the activities on a Kunjani deck, with their suit, text, answer key, assessment notes, and linked outcomes. Read-only.
publish_activity_batch
ChatGPTCreate or update a Kunjani deck and publish a batch of training activities into it in one call. When deck.id targets an existing deck, supplied deck fields (name, description, visibility, collaborations, dice_option, picture_url) are applied and omitted fields are left untouched. mode "publish" persists; mode "validate" only previews and writes nothing (no deck, outcome, or activity is created or updated). Author using the Agent K method: (1) design 3-6 learning outcomes first (specific, observable, learner-facing, e.g. "Handle an unhappy customer without escalation") and tag every activity with the outcome(s) it advances via its outcomes field. (2) Each activity has three parts: text is what the player sees, a scenario or trigger that must NOT teach or reveal the answer; answer is the answer key the AI grader scores against (specific, bulleted key points); assessment_notes are grading rules and context (accept synonyms and real-world examples, state the minimum valid points, do NOT write rigid tier rubrics, and include any media transcript because the grader is text-only). (3) suit sets the activity type: Jolt (quick recall, 30-60s), Advance (praise a good action then unpack its benefits, 60-90s), Mystery (creative photo/video/lateral thinking, 90-300s), Oops (a mistake scenario then unpack the consequences, 60-90s), Explain (summarise/compare/analyse, 90-120s), Demonstrate (perform, role-play, or submit media, 90-300s). Keep each activity to the one thing it tests, and OMIT each activity name so Kunjani auto-numbers it within its suit (J1, A1, M1, O1, E1, D1); only set a name to pin a specific slot. After publishing, share a deck link from the response with the user: deck.teams_url when the user is in Microsoft Teams or M365 Copilot, otherwise deck.url.
reorder_deck_questions
ChatGPTSet the display order of a Kunjani deck's activities, the order the deck overview and games use. question_order must list EVERY question on the deck; read the current order from get_deck first.
update_deck
ChatGPTUpdate one Kunjani deck: name, description, visibility, collaborations, dice option, or thumbnail. Only the deck fields you supply are changed. Use create_deck to make a new one and publish_activity_batch to add activities. dice_option "loaded" plays activities in a fixed order (default, for scaffolded decks); "random" fires a random activity per suit, for pure-recall drills. Deck thumbnails are URL-based: deck.picture_url fetches a public image (jpeg/png/gif/webp, max 16 MB) and deck.remove_picture true clears it. A deck can also carry ONE learning resource that players are sent after the welcome message: deck.resource_url fetches a pdf or mp4 (max 16 MB), or deck.resource_link points at a video link; supply one or the other, and deck.remove_resource true clears the uploaded file. Share deck.teams_url from the response with the user when they are in Microsoft Teams or M365 Copilot, otherwise deck.url.
update_outcome
ChatGPTChange one learning outcome on a Kunjani deck, including which activities it is linked to. outcome.question_ids REPLACES those links when supplied, and [] detaches all of them. Only outcomes the deck owns can be changed.
update_question
ChatGPTChange one activity on a Kunjani deck. Only the fields you supply change. The three-block rule still applies: text never reveals the answer, answer is the grader's answer key, assessment_notes are flexible grading rules plus any media transcript. outcome_ids or outcome_descriptions REPLACE the activity's outcome links when supplied. Media is URL-based (picture_url / picture_2_url are player-facing and facilitator_picture_url is response media; public image or audio, max 5 MB image / 16 MB audio or video). remove_picture, remove_picture_2, and remove_facilitator_picture clear them.
