fluent_add_closet_item
ChatGPTCreate one NEW Style closet item from a profile YOU (the host model) produced from the user's photo or description. A photo is optional but strongly encouraged: when you have a usable image, include it so the item gets a cover. With a reviewed primary Catalog image set catalog_ready=true; in ChatGPT prefer image_file for an uploaded image, or pass the exact inspected direct public HTTPS image as image_url and Fluent will copy and validate its bytes before the atomic create. When the Catalog is host-generated, also pass the exact retained source evidence through source_image_file (or its bounded data/hosted-file equivalent) in the same call. Without catalog_ready, an image is saved as the item's ordinary photo after the create; with no image at all the item is saved with photoStatus "needs_photo", and you should offer once to add a photo with fluent_set_closet_item_photo. Fluent stores and hashes the bytes, binds the reviewed Catalog to its source, surfaces possible duplicates for YOU to judge (a warning saves nothing), and returns read-after-write proof plus an exact reviewHandoff. This tool only creates: for a garment that is already saved, add its photo to that item with fluent_add_closet_item_photo (or fluent_set_closet_item_photo with image_type "primary" when it has no photo yet) instead of creating it again. In UI-capable hosts, a successful create directly presents the exact saved item in the current Closet app; the reviewHandoff is the text-only/non-UI fallback. A created item is active; ingestion_review is a render presentation mode, not a pending_review lifecycle status. Fluent does not inspect images, browse or scrape product pages, resolve product galleries, generate images, or infer taste; you own the visual judgment. Save only clothes the user asked you to add, told you they bought, or that their standing rule covers; the approval field says which.
fluent_add_closet_item_photo
ChatGPTAdd one more original photo (a product or detail photo, or an on-you fit photo) to a saved closet item, with read-after-write proof. Every saved photo and the current cover are kept; repeating the same image and role reuses the photo already added. In ChatGPT, pass an uploaded or selected image through image_file. Fluent copies the image into its own storage, stores and hashes the bytes, and never inspects images or browses product pages. Verify the returned photoId. To change the cover or a role slot, use fluent_set_closet_item_photo.
fluent_archive_closet_item
ChatGPTArchive one closet item the user no longer has, with an explicit reason, disposition, provenance and read-after-write proof. Use for returned, sold, donated, gifted, worn-out, never-purchased, cancelled, or otherwise gone items. The item leaves the active closet; audit history is kept and fluent_restore_closet_item brings it back. This tool only archives: to combine a duplicate record into the item that stays, use fluent_merge_closet_items. Meals is retired; domain="meals" archives nothing.
fluent_get_account
ChatGPTReturn the account profile represented by the authenticated connection for account labeling. The opaque ID remains stable across reconnects, token refresh, and display-name changes.
fluent_get_account_status
ChatGPTFetch the data-minimized Fluent account/status surface for account access, export, deletion, reactivation, and support. Returns access state, enabled domains, account and support links, export and deletion instructions, and support email. It does not start, sell, upgrade, cancel, or manage paid access from the assistant.
fluent_get_capabilities
ChatGPTFetch backend mode, contract version, available domains, enabled domains, onboarding state, and starter workflow discovery hints for Fluent tool routing.
fluent_get_closet_context
ChatGPTFetch a compact context packet for a domain and intent. Use domain="style" with intent="closet" before naming clothes the user owns in an outfit or "what should I wear?" answer. The host model owns reasoning and final judgment; Fluent supplies durable context, typed items, evidence gaps, freshness, and suggested writeback boundaries. Meals is retired; domain="meals" returns only a retirement notice.
fluent_get_closet_item
ChatGPTFetch one typed domain item with its canonical payload and provenance hooks. Use the stable item ID returned by fluent_list_closet_items; do not invent IDs. Meals is retired; Meals items return only a retirement notice.
fluent_get_closet_item_photos
ChatGPTFetch host-inspectable photos and their constraints for saved closet items. Fluent provides photo provenance and delivery; the host model must inspect images before making visual claims. For visual outfit or shoe advice in ChatGPT, call this separately for each shortlisted saved item using subject or a singleton item_ids array so the focused item can include model-visible pixels; multi-item bundles are references-only. Never rank visual color, proportion, texture, or silhouette from item text or image URLs alone.
fluent_get_profile
ChatGPTFetch the shared profile envelope: core profile facts, capabilities, boundaries, and provenance-ready fact slots. Domain-specific payloads stay typed and are not flattened into generic memory.
fluent_hide_closet_item_photo
ChatGPTHide one visible photo of a saved closet item. The photo file is kept, not deleted, so the change can be undone. Hiding the current cover requires next_cover_photo_id. Send the item's photoRevision from fluent_show_closet as expected_revision; a changed revision fails without saving. Returns read-after-write proof and an undoToken for fluent_undo_closet_item_photo_change. For user-approved closet management only.
fluent_list_closet_evidence
ChatGPTList evidence, provenance, source snapshots, event history, or evidence gaps for a subject or claim. Evidence can inform host reasoning, but it is not a final plan, style verdict, medical judgment, or checkout action.
fluent_list_closet_items
ChatGPTList typed domain items such as Style closet items. Results are incomplete whenever nextCursor is returned: never conclude that an item is absent until relevant pages are exhausted or a confident match is found. Prefer targeted query values before broad closet traversal. For Style photo ingestion, search discriminating brand or graphic text, garment description, and category/color evidence before creating; a matching item without Catalog media is an incomplete existing item to repair, not a new item. For owned-item outfit or shoe advice, list the relevant active Style category before recommending a saved item, then visually inspect shortlisted items through focused fluent_get_closet_item_photos calls.
fluent_merge_closet_items
ChatGPTMerge a duplicate closet record into the item that stays, after the user confirms they are the same physical garment. Fluent moves the duplicate's retained photos into the kept item, fills only its missing core details, and archives the duplicate, with read-after-write proof. Never infer or auto-merge. The result carries the merge_id that fluent_undo_closet_item_merge needs. For user-approved closet management only.
fluent_record_closet_item_feedback
ChatGPTRefresh selected profile fields or typed natural-language wear feedback on an EXISTING saved Style closet item from explicit host/user evidence. This is the durable home for tags, descriptors, fit/wear learning, and recommendation corrections. Fluent rank-merges each field, preserves stronger existing evidence, never infers unworn from silence, and returns read-after-write proof.
fluent_record_closet_view
ChatGPTCalled by the Fluent closet view, not by the assistant: records which page, item or pieces the person has open so a later fluent_get_closet_context can say what they are looking at. Changes nothing in the closet.
fluent_reorder_closet_item_photos
ChatGPTSet the display order of a saved closet item's visible photos. Send the item's photoRevision from fluent_show_closet as expected_revision; a changed revision fails without saving. Returns read-after-write proof and an undoToken for fluent_undo_closet_item_photo_change. For user-approved closet management only.
fluent_replace_closet_item_photo
ChatGPTSwap one visible photo of a saved closet item for a photo already added to the same item (for example with fluent_add_closet_item_photo). The replacement takes the old photo's place, and the cover if it was the cover; the old file is kept. Send the item's photoRevision from fluent_show_closet as expected_revision; a changed revision fails without saving. Returns read-after-write proof and an undoToken for fluent_undo_closet_item_photo_change. For user-approved closet management only.
fluent_restore_closet_item
ChatGPTRestore one archived closet item to the active closet, with read-after-write proof. Use it when the user says an archived item is back or was archived by mistake. It does not undo a duplicate merge: an item archived by a merge is rejected without changes and must use fluent_undo_closet_item_merge. For user-approved closet management only.
fluent_save_closet_item_product_details
ChatGPTSave the product YOU (the host) identified for one saved closet item, with read-after-write proof: the product listing, its match status and basis, and attributed facts (colour, composition, care, construction, made in) that each cite a source. The details replace the previous ones in full, so carry forward every existing fact you keep. Read productReference.revision from fluent_get_closet_item and send it as expected_revision (0 if absent). Keep uncertain matches as candidate until the user confirms. Fluent does not browse or fetch the listing, change photos or edit ownership fields. For a new item, add it first, then save its product details. For user-approved closet management only.
fluent_set_closet_item_cover
ChatGPTMake one visible product photo the cover of a saved closet item. On-you (fit) photos cannot be the cover. Send the item's photoRevision from fluent_show_closet as expected_revision; a changed revision fails without saving. Returns read-after-write proof and an undoToken for fluent_undo_closet_item_photo_change. For user-approved closet management only.
fluent_set_closet_item_photo
ChatGPTSet the photo in one role slot of a saved closet item (primary cover, alternate or fit), replacing what that slot held, with read-after-write proof. In ChatGPT, pass an uploaded, selected, or generated image through image_file. Set catalog_ready=true only after reviewing the final primary Catalog presentation; for host-generated media, also provide source_photo_id for the exact retained non-generated source on this item. Fluent stores and hashes inline/hosted-file bytes and atomically binds the reviewed Catalog to its source. It never inspects images or browses product pages. To add another photo while keeping every saved photo and the cover, use fluent_add_closet_item_photo.
fluent_show_closet
ChatGPTShow the Fluent Closet of the user as an MCP Apps view. Reads owned Style items and returns one collection-to-detail experience plus compact structured fallback data. When the user asks to review, compare, or resolve a possible duplicate between two saved closet records, call this tool with exactly those two saved IDs in filter.item_ids, filter.status="active", and presentation.mode="comparison"; omit focused_item_id. Do not claim that the comparison is open unless this tool was actually called and returned. After an outfit or shoe recommendation names one winning owned item, render the conversational payoff with filter.item_ids containing only its exact saved ID, presentation.mode="recommendation", focused_item_id set to the same ID, and one concise recommendation_reason; never substitute a category/subcategory/query filter or the whole closet. For an outfit of 3 to 5 exact owned items, render only those selected saved IDs with presentation.mode="comparison" and omit focused_item_id; the model must name the exact items and may offer at most one exact-owned replacement. Use presentation.mode="ingestion_review" only as the text-only/non-UI fallback when fluent_add_closet_item could not attach its create-owned review app: follow fluent_add_closet_item payload.reviewHandoff with filter.status="active", exact filter.item_ids, and the exact focused_item_id. Newly created items are active; pending_review is not a valid lifecycle or filter status. Exact filter.item_ids otherwise default to comparison and always provide an in-app route back to the full closet. Use mode="detail" with focused_item_id only when the user explicitly asks to inspect or manage the complete saved item. Category/type/brand/color/size/query filters are for deliberate collection browsing; status="archived" shows archived items. This surface manages saved closet state and hands exact stored item IDs to outfit creation; it does not invent garments or make purchase judgments.
fluent_undo_closet_item_merge
ChatGPTUndo one duplicate merge, with read-after-write proof: the archived duplicate becomes active again and the photos it gave the kept item move back to it. Send the archived item's ID and the exact merge_id of that merge (duplicateMergeId on the item in fluent_show_closet). If that item has no outstanding merge with this merge_id, nothing changes and the call fails. For user-approved closet management only.
fluent_undo_closet_item_photo_change
ChatGPTReverse the most recent cover, order, hide or replace change on a saved closet item's photos, using the undoToken that change returned. It only works while the photos are unchanged since. Send the item's photoRevision from fluent_show_closet as expected_revision; a changed revision fails without saving. Returns read-after-write proof and an undoToken for fluent_undo_closet_item_photo_change. For user-approved closet management only.
fluent_update_closet_item
ChatGPTEdit the catalog details of one saved closet item (brand, name, category, subcategory, color, size, formality) with read-after-write proof. Omitted fields stay unchanged. This tool only edits details: tags and styling descriptors belong in fluent_record_closet_item_feedback, and restoring an archived item uses fluent_restore_closet_item. For user-approved closet management only; Fluent does not browse retailers or read email.
fluent_update_profile
ChatGPTSave one explicit, provenance-backed fact about the user (for example their timezone, a size note or a style preference) through the canonical shared or domain profile service. Use only when the user intends to change durable Fluent memory.
complete_task
Claudecreate_person
Claudecreate_task
Claudeget_daily_briefing
Claudeget_meeting_notes
Claudeget_memory
Claudeget_my_day
Claudeget_person_context
Claudeget_recent_commitments
Claudelist_follow_ups
Claudelist_goals
Claudelist_memories
Claudelist_people
Claudelist_tasks
Claudesearch_conversations
Claudeset_day_plan
Claudeupdate_person_notes
Claude