mcp server

designforyou

Generate logos, social posts, app screenshots, comic panels & visual-novel assets from prompts.

Description as published by the maintainer. Source

• version 1.0.1
• active

active — Registry entry last updated 2026-06-18.

What this server can do

41 functions, named and described by the server itself. Parameter names are shown because they say more about what a function does than its name usually does.

animate_reference_grok_video(line, model, prompt, duration, image_url, aspect_ratio, image_data_url)

Generate a short image-to-video take with xAI Grok Imagine Video from one reference frame. The result is normalized to the requested source duration. Required: prompt.

animate_reference_image(line, model, prompt, duration, image_url, aspect_ratio, batch_approved, image_data_url)

Generate a short vertical image-to-video take from one supplied reference frame. Use an HTTPS artifact URL or a data URL when exact source style and composition need to anchor the shot. Required: prompt.

animate_reference_video(line, model, prompt, duration, video_url, aspect_ratio, batch_approved, video_data_url)

Generate a short video take conditioned on one supplied source clip. Preserves the source shot's actors, illustration style, wardrobe, camera, and action more closely than a still-frame reference. Required: prompt.

assemble_shortform_reel(fps, width, height, segments, output_path, segment_urls, audio_data_url, segment_durations, segment_data_overrides)

Assemble ordered local video segments or HTTPS generated-artifact URLs into one normalized 9:16 MP4. This is the missing composition step for multi-take LetMeActForYou stories.

audition(cue, line, emotion, actor_id, casting_id, batch_approved, direction_notes)

Start a casting audition: the chosen actor performs your cue in a short (~8-10s) vertical video. Billed. Returns audition_id immediately; poll audition_status every few seconds until status=ready, then share video_url. Get actor_id from casting_roster. Required: cue.

audition_status(audition_id)

Check a casting audition started with audition: status (rendering / ready / failed) and, when ready, the video_url to share. Free to call. Required: audition_id.

browse_templates(category, subcategory)

Browse the DesignForYou template catalog. Optionally filter by category (e.g. 'instagram', 'app-store', 'logo', 'living-panels'). Returns template IDs, names, and previews.

cast_character(notes, actor_id, story_id, source_url, character_name, character_description)

Bind a seeded or private actor to one story character. Returns a casting_id that keeps character identity and story provenance across takes. Required: actor_id, story_id, character_name, character_description.

casting_roster(castingrosterinput)

List only the public seeded LetMeActForYou casting roster with headshots, type, and bio. This public catalog never includes a user's private actors; authenticated users can retrieve those with my_actors. Call this first, show the user the seed actors, then call audition with the chosen actor_id.

comic_design_brief(premise)

Return the no-server-LLM Living Panels authoring brief: PanelScript instructions, schema, unit templates, and success criteria. Use this when the MCP client will do the creative direction with its own model, then call comic_submit_design with the authored Story JSON. Required: premise.

comic_revision_brief(comic_id)

Return the current Story JSON, stale-write hash, allowed patch paths, and dirty-node contract for a client-authored Living Panels revision. Use this when the client will decide the edit with its own model, then call comic_submit_revision. Required: comic_id.

comic_status(comic_id)

Check asset-production progress for a comic created with create_comic: panels, motion clips, and voice lines done vs total, plus ready=true when the whole comic is live at its reader_url. Free to call. Required: comic_id.

comic_submit_design(story, premise, client_name, client_model, start_production)

Accept a client-authored Living Panels Story JSON and start artifact production without using a server story LLM. The client does direction; DesignForYou validates, saves, generates character sheets, panel/unit art, motion clips, voice-over, and assembles the public reader. Required: premise, story.

comic_submit_revision(story, changes, summary, comic_id, client_name, dirty_nodes, client_model, base_story_hash, start_production)

Apply a client-authored Living Panels revision without using a server revision LLM. Targeted changes regenerate only inferred dirty producer nodes such as panel:<id>, motion:<id>, or tts:<bubble_id>. Required: comic_id, base_story_hash.

create_actor(bio, name, emoji, actor_type, visual_description, headshot_template_id, performance_reference_url)

Create a private LetMeActForYou actor draft. Generates a casting headshot, stores it on S3, and returns actor_id for cast_character or audition. Draft actors never appear in the public feed; an authorized operator can call publish_actor with an approved seamless idle-loop URL when the actor is ready. When available, provide a saved full-body standing performance_reference_url; all generated clips prioritize it over the profile headshot. The default casting template produces an original Pixar-quality 3D animated feature-film headshot; pass stable fictional physical identity, not performance direction. Apparent age may be child, teen, adult, or elderly. Required: name, visual_description.

create_comic(premise)

Write and produce a Living Panels interactive comic from a one-paragraph premise. One call writes a complete 5-page comic (panels, dialogue, motion plan, voice casting) and starts generating every asset in the background: character sheet, ~13 panels of art with character consistency, ambient image-to-video motion for the liveliest panels, and per-line voice-over. Returns immediately with the story summary and a public reader_url where the comic can be read, watched live, and heard. Production takes several minutes — poll comic_status until ready, then share the reader_url. Required: premise.

create_template(name, width, height, best_for, category, base_prompt, subcategory, text_fields, aspect_ratio, layout_rules)

Create a private reusable DesignForYou template from a structured brief. The returned template_id works immediately with get_text_fields and generate. This stores a user-owned template in the canonical catalog collection; it does not publish it to the anonymous gallery. Required: name, category, aspect_ratio, width, height, base_prompt, layout_rules, text_fields.

delete_actor(actor_id)

Delete an unused private actor and headshot. Refuses seeded actors and actors referenced by active castings or auditions. Destructive and free. Required: actor_id.

delete_audition(audition_id)

Permanently remove the authenticated user's media and user-authored direction for one ready or refunded-failed audition. Keeps a minimal tombstone for billing, refund, and lineage audit. Destructive and free. Required: audition_id.

delete_casting(casting_id)

Delete an unused authenticated-user casting. Refuses while active auditions reference it. Destructive and free. Required: casting_id.

extend_minimax_h3_video(prompt, duration, source_request_id)

Queue one paid MiniMax H3 continuation from a completed video owned by the authenticated caller. The continuation starts from the source video's final frame, preserves eligible references, and costs 5 credits. Duration must be 5, 10, or 15 seconds. Returns the new queued request ID and lineage metadata including source_request_id, generation_number, and total_duration. Required: source_request_id, prompt, duration.

generate(model, variants, edit_tier, session_id, context_ids, description, template_id, context_labels, skip_prompt_merge)

Generate one or more finished images from a template (get a template_id from recommend_template or browse_templates) plus a description. Use this after the user has selected or explicitly supplied a template_id; otherwise call `recommend_template` first so the visual gallery can collect the selection. Works for all categories (Instagram, logo, app-store, Visual Novel backgrounds/sprites, CG illustrations). Pass variants for multi-image output (expressions, time-of-day, etc.). Pass session_id to refine a prior result. Pass context_ids to ground a new image on prior designs (character consistency for VN CG scenes). Required: template_id, description.

generate_character_video(beats, style, duration, location, character, aspect_ratio, prompt_model, requirements, batch_approved)

Generate a short realistic character video from structured requirements. The tool gives DesignForYou's reusable home-video character template and the user's requirements to a text LLM, which writes a fresh final video prompt while preserving character consistency, camera imperfections, ambient audio, and requested shot structure. It then generates a Gemini Omni Flash video. Use this for 3-10 second character/lifestyle clips.

generate_endless_idle_loop(hq, prompt, duration, provider, image_url, resolution, euler_steps, aspect_ratio, image_data_url)

Create a silent looping idle video from one reference image. MiniMax H3 receives two byte-identical image artifacts explicitly tagged as first and last frames. The native signed-in Grok CLI path additionally replaces the delivered first and last frames with the source image after generation.

generate_scene_frame(mood, pose, model, location, provider, wardrobe, actor_ids, composition, aspect_ratio, composition_reference_url, continuity_scene_frame_id)

Generate one actor-grounded narrative first frame from one to four existing LetMeActForYou actor_ids plus location, wardrobe, pose, and composition. Uses actor headshots as image references and returns a stored image_url. For multi-shot continuity, first generate a wide scene reference, then pass its scene_frame_id as continuity_scene_frame_id while specifying a fresh camera setup in composition and fresh blocking in pose for every shot. Pass only the actor_ids that should be visible in that shot. Pass composition_reference_url to preserve a Blender previs frame's camera and blocking while replacing proxy geometry with actors and the finished set. Defaults to Gemini; pass provider='grok' to use Grok Imagine. Use this before animation when a close-up headshot is not enough. Required: actor_ids, location.

get_credits

Return the authenticated user's current credit balance, usage, and Stripe subscription status. Free to call.

get_text_fields(template_id)

List a template's editable text fields (headline, subhead, CTA, etc.) with their current values. Free. Use right after a template is chosen and BEFORE `generate` when you want to fill the template's fields precisely instead of relying on a freeform description. Required: template_id.

minimax_h3_video_status(request_id, wait_seconds)

Check a queued MiniMax H3 video for free. Pass wait_seconds (up to 900) to keep this MCP call open until the job completes, fails, or is cancelled; completed jobs include the persistent S3 video_url. This works from any MCP thread authenticated as the same account that queued the job. Required: request_id.

minute_clip_composer(delta, action, job_id, clip_id, dialogue, segments, story_brief, render_provider, production_bible)

Create and advance one shared twelve-clip, five-second composer job. The same parent job is visible in the website UI and in MCP history. A client model should call planning_scaffold first, then author and submit the four-section production_bible and twelve continuity-aware segments itself. Use create with a story_brief, then build_bible without production_bible to run the story-brief agent (or supply an explicit production_bible), then build_shot_plan without segments to run the shot-plan agent (or supply explicit segments), then open_generation, and generate_clip as the agent completes each stage. cancel_clip restores the exact pre-generation shot state. generate_clip defaults to MiniMax H3; pass render_provider='grok_cli' to use the signed-in Grok sidecar with first-frame handoffs. Pass plain dialogue lines of no more than eight words with speaker and voice_context; the renderer applies provider dialogue tags in the assembled prompt. Required: action.

my_actors(limit, offset)

List only the authenticated user's private reusable LetMeActForYou actors, newest first. Free. Supports bounded offset pagination and never returns internal identity prompts or visual descriptions.

my_auditions(limit, offset, status)

List only the authenticated user's LetMeActForYou audition history, newest first. Free. Supports bounded offset pagination and an optional queued, rendering, ready, or failed status filter.

my_minute_clip_jobs(limit, offset)

List the shared parent composer jobs created from MCP or the DesignForYou video UI, newest first.

publish_actor(order, actor_id, idle_video_url)

Promote an operator-owned actor draft into the shared public LetMeActForYou roster and bind its approved seamless idle-loop video. Only configured public actor publishers and first-party internal services can call this tool. Auditions remain private to the user who generated them. Required: actor_id, idle_video_url.

queue_minimax_h3_videos(hq, prompts, duration, image_urls, resolution, euler_steps, image_roles, aspect_ratio)

Add one or more prompts to your private MiniMax H3 service queue. Each prompt costs 5 credits. Choose 5, 8, 10, or 15 seconds, and Euler sampling steps of 8, 15, or 20. Set hq=true for approximately 1MP output (1344x768 landscape or 768x1344 portrait). Choose 16:9 landscape or 9:16 mobile portrait output (480x864 or 768x1344). Optionally attach up to five image URLs; images are references by default, with optional first/last frame roles. Queued jobs resume automatically when the service returns online. Required: prompts.

recommend_template(needs, category, use_case, subcategory)

Recommend the best matching templates for any use case. Describe what you need and the system auto-detects the category and ranks templates by relevance. Return the recommendations so the user can pick a template from the visual gallery; do not call `generate` in the same turn unless the user explicitly names a template_id or asks you to skip selection. Required: use_case.

remove_minimax_h3_queue_job(request_id)

Remove one queued, failed, or cancelled MiniMax H3 job. For an active render, use the safe cancellation action instead. Required: request_id.

reorder_minimax_h3_queue(direction, request_id)

Move a queued MiniMax H3 job one place earlier or later. It never interrupts an active render. Required: request_id, direction.

report_casting_abuse(reason, details, target_id, target_type)

Report a visible LetMeActForYou actor or an audition owned by the authenticated user. Free. Stores bounded triage metadata without copying prompts, media URLs, billing data, or account identity. Required: target_type, target_id, reason.

revise_audition(cue, line, emotion, batch_approved, direction_notes, parent_audition_id)

Create a child take from an existing audition. Inherits actor, story casting, cue, emotion, and line unless overrides are supplied, while preserving parent/root ids and the director note. Required: parent_audition_id, direction_notes.

vn_generate_sprite_sheet(cells, model, sheet, style, output, quality, provider, character)

Generate one fixed-grid visual-novel character sprite sheet, crop it into cell assets, remove the chroma-key background, validate transparent sprites, and return a VN Creator-ready manifest. Base cost: 3 credits. Optional quality.hi_res_cells=true adds a reference-grounded per-cell re-render at full native resolution (fixes cells being limited to ~1/N of the sheet resolution) and bills +2 credits per cell re-rendered. quality.hi_res_model optionally routes the hi-res pass to a specific model (e.g. gpt-image-2 for fabric fidelity). Required: character, style, sheet, cells.

widget_diagnostic(mode)

Return a fixed, no-cost DesignForYou widget payload for ChatGPT Apps debugging. Use this only to verify the template gallery/result iframe renders correctly; it does not generate images, use an LLM, or charge credits.

Last successful function declaration observed on 2026-08-13. Source: https://designforyou.swapp1990.org/mcp/v2/. We list what the server declared; we do not call any of these functions.

Endpoint status observed on 2026-08-13. Source: https://designforyou.swapp1990.org/mcp/v2/.

Signals

These are separate measurements of different things. They are deliberately not combined into one score, because a popularity number that mixes website traffic with saves and stars cannot be checked or acted on.

Where to get it

• Repository — https://github.com/swapp1990/designforyou-mcp
• Website — https://designforyou.swapp1990.org
• Remote endpoint (streamable-http) — https://designforyou.swapp1990.org/mcp/v2/