create_workflow
ChatGPTCreate a new Workflow canvas (a reusable multi-step generation pipeline) from a step plan. The server compiles the plan into nodes and wiring, saves it, and the Workflow page in the VidGuy app opens it. Free: nothing runs and no credits are spent; the user runs it on the canvas, and estimatedCredits is what a full run would cost. Use it when the user wants a pipeline that chains steps (for example: concept prompt → 3 image variations → animate the best → voiceover), a repeatable recipe, or says "build a workflow". For one image or video, use the Studio composer instead. Steps: give every step a short stable id; kind is prompt | image | video | voice | music | audio | quickCreate | reference. inputs wire earlier steps: [{fromId, as: text|image|video|audio}]. image accepts text/image; video accepts text/image/video; voice accepts text; music accepts text/video; audio accepts text/image/audio; quickCreate accepts text/image/video. Never wire audio into video or quickCreate. A user-supplied image is a reference step with options.url (https) and no inputs, wired as image into every step that must match it. Don't add an assemble step. model is an optional Studio model key (defaults are picked). options.label sets the node label.
estimate_short_form_post_pack_credits
ChatGPTEstimate the credits needed to generate a VidGuy pack of short-form video and image posts. This only calculates a price; it does not generate content or spend credits.
generate_meme
ChatGPTCreate memes from a plain-language prompt — a pasted changelog line, blog title, or launch note works; you do not need to write the joke. Choose meme-edit (rewrites a familiar reaction clip, 3 credits each), meme-greenscreen (keys a person over a background with a hook, 2 credits each), or meme-comic (an illustrated comic strip, 2 credits each). count accepts 1–5 and produces that many variations of the selected type. brandId is OPTIONAL: omit it for a brandless meme driven purely by the prompt, or pass one from list_brands to fold in that brand's voice. ASYNC: returns a jobId immediately; poll get_job, and expect generation to take ~1–3 minutes. Credits refund automatically on failure. Individual slots can silently skip — especially meme-edit, which needs a source clip whose caption rewrites cleanly — so fewer than count memes may land.
get_brand
ChatGPTFetch a single brand by its id.
get_credits
ChatGPTCheck the account's current credit balance, plan allotment, and recent credit transactions. Call this before a credit-spending generation (create_video, generate_snacks, generate_meme, studio_, seedance_, create_slides/ai_story, post-processing) to confirm there's enough balance. Read-only; no credits spent.
get_job
ChatGPTPoll ANY async VidGuy job by its id — the universal poller. Covers: jobIds from generate_snacks / generate_meme / create_video / create_slides / create_ai_story / create_tiktok_carousel (whose slide URLs come back on the job itself, no second call), Studio + SeedDance generations (jobId from studio_generate_video / seedance_generate_video / seedance_generate_image), and AI Edit / Character Swap projects (projectId from create_ai_edit / create_character_swap). status is queued|processing|completed|failed (Studio/SeedDance report 'succeeded' when done). Generations take ≥~2 minutes — poll every ~10s and don't treat an early 'processing' as a failure. TikTok Carousels are the exception: they paint in seconds, so poll those every ~3s. Completed outputs (videos, images, snack dashboard links) are attached as resource links so they display inline. Only exceptions: influencer vault builds / snack packs (get_influencer_vault_build / get_influencer_snack) and dubs (get_dub_status) have dedicated pollers.
get_short_form_post
ChatGPTGet one previously generated VidGuy short-form post by ID, including its project details, source visuals, and dashboard URL.
get_video
ChatGPTFetch one video project by its id: status, outputUrl, the brief, and per-format assets (slides tiles). The output video is attached as a resource link when available. For a Character Swap project it also returns characterSwap — the exact settings it ran with (r2Keys, keepOriginalVisualRanges, resolution, sceneMatch, saveAudio) plus signed previews — so you can replay the same cut with a different character.
get_vidguy_studio_snapshot
ChatGPTRefresh the VidGuy Studio balance and first library page. Read-only; no credits spent.
get_workflow
ChatGPTRead one Workflow canvas: every node (id, type, label, prompt/model settings, run status, output URL) and every edge with its handles, plus the revision to pass to the workflow editing tools. Omit graphId for the most recently edited workflow. When the user has the Workflow page open in the VidGuy app, its graphId is in your context. Read-only; no credits spent.
get_workflow_guide
ChatGPTThe Workflow canvas reference: node types, their settings, valid port handles and wiring rules, and editing conventions. Read it once before creating or editing a workflow in a conversation. Read-only; no credits spent.
list_brands
ChatGPTList the brands (end-customers) registered under your account. Paginated. Call this first to find a brandId before generating snacks.
list_short_form_posts
ChatGPTList previously generated VidGuy short-form video and image posts, with optional brand filtering and pagination.
list_studio_models
ChatGPTList the Studio image/video/audio model catalog (keys, pricing, capabilities, min plan). Pass a model's key to studio_generate_image/video/audio.
list_videos
ChatGPTList your video projects, newest first. Use this to find a completed project's projectId for post-processing (upscale_video / voice_enhance / dub_video) or to check generation status. Filter by status (e.g. 'completed'). Paginated.
list_workflows
ChatGPTList the user's VidGuy Workflow canvases (node graphs), most recently edited first. Read-only; no credits spent.
open_vidguy_studio
ChatGPTOpen the VidGuy app: the user's Studio library of generated images, videos and audio with a composer to make more, plus their Library, Quick Create, Workflow canvas, Seedance Studio and Brand Hub pages. While the app is open, it tells you what the user is looking at (an item, a page, or the Workflow canvas with its nodes). Read-only to open; nothing is generated and no credits are spent until the user starts a generation.
studio_app_generate
ChatGPTStart an image, video, or audio generation from the Studio composer after the user presses Generate. SPENDS CREDITS (the composer shows the cost first). Images and audio return finished; videos return a jobId to poll with studio_app_job.
studio_app_job
ChatGPTCheck a Studio video generation started from the composer. Read-only apart from finalizing a finished job; no new credits spent.
studio_app_library
ChatGPTLoad one page of the user's VidGuy library for a Studio tab. Read-only; no credits spent.
studio_app_models
ChatGPTList the image, video, and audio models the user can run from the Studio composer, with credit pricing. Read-only; no credits spent.
studio_app_open_page
ChatGPTOpen a VidGuy website page (Library, Quick Create, Workflow, Seedance Studio, Brand Hub, Accounts, or one video, snack pack, editor project, or workflow graph by id) inside the Studio app. With handoff, the link signs the app's frame in once. No credits spent.
studio_generate_image
ChatGPTGenerate a single image via the Studio model catalog (call list_studio_models for keys/pricing). SYNC — the asset URL comes back in the result. SPENDS CREDITS (per-model, see creditCost). Reference images can be passed as URLs (use import_asset_from_url / create_upload first if needed). negativePrompt applies only to models whose capabilities include supportsNegativePrompt (e.g. NOT BytePlus/Seedream); on other models it is ignored, not an error.
studio_generate_video
ChatGPTGenerate a single video clip via the Studio model catalog (call list_studio_models for keys). ASYNC — returns a jobId; poll get_job until status is succeeded/failed. SPENDS CREDITS. firstFrame/referenceImages accept URLs (import them first if needed).
workflow_add_node
ChatGPTAdd one prompt, generation, output, or comment node to an existing Workflow canvas. Returns its nodeId for later connections. If x/y are omitted, the whole canvas is automatically laid out. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_add_reference_node
ChatGPTAdd one reference image or reference video node using a user-provided media URL. Stores the URL without fetching or generating media. Returns its nodeId. If x/y are omitted, the whole canvas is automatically laid out. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_clear_canvas
ChatGPTDelete all nodes and connections from an existing Workflow canvas. Use only after the user explicitly requests starting over, with confirm:true. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_connect_nodes
ChatGPTConnect one output handle to one compatible input handle using existing node IDs. For a single-input handle, replaces its existing connection. Does not run either node. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_disconnect_edge
ChatGPTRemove one connection by its edgeId. Both nodes and their settings and outputs are kept. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_move_node
ChatGPTMove one existing node to canvas coordinates x/y, replacing its previous position. Keeps its configuration and connections. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_remove_node
ChatGPTDelete one node and all connections touching it from the Workflow canvas. Removes its saved settings and canvas output references. Use only when the user requests removing that node. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_set_project_name
ChatGPTReplace the Workflow canvas name with the supplied name. Keeps all nodes, connections, and outputs. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_tidy_canvas
ChatGPTRearrange the existing nodes into a left-to-right layout, replacing their positions. Keeps all nodes, settings, connections, and outputs. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.
workflow_update_node
ChatGPTPatch one existing node's configuration, such as its prompt, model, duration, or aspect ratio. Overwrites the supplied settings; other settings and existing outputs are kept. config must contain at least one field. The Workflow page open in the app updates live. Free: no node runs and no credits are spent. Pass expectedRevision from get_workflow or the last edit result; on conflict, read get_workflow and retry. Node IDs must be actual IDs, not @ref aliases.