Search / com.brix36/brix36
Brix36
R0· no sign-inFilms and music videos. You're the star. No subscription.
v0.2.0 · active · website · descriptor JSON · versions
Search / com.brix36/brix36
R0· no sign-inFilms and music videos. You're the star. No subscription.
v0.2.0 · active · website · descriptor JSON · versions
{
"mcpServers": {
"brix36": {
"type": "http",
"url": "https://api.brix36.com/mcp"
}
}
}claude mcp add --transport http brix36 https://api.brix36.com/mcp[mcp_servers.brix36]
url = "https://api.brix36.com/mcp"
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"brix36": {
"type": "remote",
"url": "https://api.brix36.com/mcp",
"enabled": true
}
}
}{
"mcpServers": {
"brix36": {
"type": "http",
"url": "https://api.brix36.com/mcp"
}
},
"deeplink": "cursor://anysphere.cursor-deeplink/mcp/install?name=brix36&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vYXBpLmJyaXgzNi5jb20vbWNwIn0%3D"
}{
"servers": {
"brix36": {
"type": "http",
"url": "https://api.brix36.com/mcp"
}
}
}{
"mcpServers": {
"brix36": {
"httpUrl": "https://api.brix36.com/mcp"
}
}
}extensions:
"brix36":
enabled: true
name: "brix36"
type: streamable_http
uri: "https://api.brix36.com/mcp"
timeout: 300
protocol.modernsupports 2026-07-28 (server/discover)protocol.statelessanswers without a sessionprotocol.transportstreamable HTTPprotocol.list_ttltools/list declares ttlMs 0, so clients re-fetch it on every useauth.prm, auth.as_metadata, auth.cimdno remote uses OAuthauth.secret_in_urlno templated URLtools.descriptionsevery tool has a descriptiontools.description_lengthdescriptions are concisetools.schemasevery tool has an object input schematools.annotations115 of 115 tools declare readOnlyHint or destructiveHinttools.directory_hintsevery tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHinttools.token_costabout 36,182 tokens to load every tool, a large share of an agent's contexttools.api_dumptools are not a one-to-one API dumpstability.changes152 tool changes in 30 daysstability.rug_pull2 tools kept their name but changed most of their description in 30 days; review before trusting themdeps.known_vulns, deps.mcp_sdk_version, deps.resolvableno npm or PyPI packagestudio_artist_ready_made_choosechangeddescriptionSaves one Brix36 ready-made fictional portrait on an existing artist using the same service as the app. Requires artist permission and a starId from studio_ready_made_stars_list. Returns the saved reference pack and whether it was newly created; choosing the same star retains the existing pack. No image generation or credit charge. Current generation readiness is read separately with studio_artists_list.
Saves one Brix36 ready-made fictional portrait on an existing artist using the same service as the app. Requires artist permission and a current starId from the ready-made star list. Returns the saved reference pack and whether it was newly created; choosing the same star retains the existing pack. No image generation or credit charge. Current generation readiness is read separately from the artist list.
studio_ready_made_stars_listchangeddescriptionLists the same free fictional character portraits offered by the Brix36 app, with each current star ID, name, description and versioned preview image. Read-only; creates no artist, image, generation or charge. Choose a portrait for an existing artist with studio_artist_ready_made_choose.
Lists the same free fictional character portraits offered by the Brix36 app, with each current star ID, name, description and versioned preview image. Read-only; creates no artist, image, generation or charge.
studio_ai_sharing_choosechangeddescriptionRecords the person's own answer to the AI-sharing question after a creation tool returned ai_processing_choice_required. Call it only with their explicit reply: allowed true for Allow, false for Not now. Never infer consent from the request to create, and never answer on their behalf. Pass supersedes as the sequence that refusal returned (null when it was unset). Returns the stored answer; nothing is created or charged.
Records the person's own answer to the AI-sharing question after a creation tool returned ai_processing_choice_required. allowed true means Allow and false means Not now. Consent counts only as that explicit reply; it is never inferred from the request to create and never supplied on the person's behalf. supersedes is the sequence that refusal returned (null when it was unset). Returns the stored answer; nothing is created or charged.
studio_motion_catalogueaddedList available Motion Graphics templates (lyric_video, audio_visualiser, living_cover, kinetic_type, club_tunnel, organic_landscape), 7 genre looks (pop, hiphop_drill, rnb, house_edm, rock, country_acoustic, afrobeats), and all 24 popular motion styles with preview video URLs and input requirements. Read-only discovery.
studio_motion_quotechangeddescriptionReturns the server price in credits and GBP including VAT, bound to the exact revision, output, source and runtime. A paid render requires one explicit customer approval of this exact price, followed by studio_motion_confirm. Edits invalidate prior quotes. No debit or render.
Returns the server price in credits and GBP including VAT, bound to the exact revision, output, source and runtime. A paid render requires one explicit customer approval of this exact price. Edits invalidate prior quotes. No debit or render.
studio_motion_createchangeddescription, input schemaSaves a ready server-rendered Motion scene and returns its exact paid-render quote plus an optional free first-frame preview. Supports lyric video, beat-reactive visualiser, kinetic titles, photo-motion poster and promo cards. Uses owned audio and picture asset IDs; saved song timing supplies lyric cues and the saved scene can be edited. No AI video model or paid render is called. requestId identifies the saved document across retries. Each returned quote has its own identity; rendering requires explicit customer approval bound to that quote.
Saves a ready server-rendered Motion scene and returns its exact paid-render quote plus an optional free first-frame preview. Supports lyric video, beat-reactive visualiser, kinetic titles, photo-motion poster, promo cards, the 6 music/lyric/visualiser templates (lyric_video, audio_visualiser, living_cover, kinetic_type, club_tunnel, organic_landscape) across 7 genre looks (pop, hiphop_drill, rnb, house_edm, rock, country_acoustic, afrobeats), and all 24 popular motion styles (styleId). Uses owned audio and picture asset IDs; saved song timing supplies lyric cues and the saved scene can be edited. No AI video model or paid render is called. requestId identifies the saved document across retries. Each returned quote has its own identity; rendering requires explicit customer approval bound to that quote.
studio_film_set_durationchangeddescriptionSet one scene’s duration, normalized to 4–30 seconds; later scenes move to follow it. Read studio_film_read first. Uses the current revision and stable operationId; retry identical inputs only. No model call or credit charge.
Set one scene’s duration, normalized to 4–30 seconds; later scenes move to follow it. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge.
studio_film_set_lip_syncchangeddescriptionSet the lip-sync choice for one saved scene. Read studio_film_read first. Uses the current revision and stable operationId; retry identical inputs only. No model call or credit charge.
Set the lip-sync choice for one saved scene. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge.
studio_film_toggle_soundchangeddescriptionToggle one Film music, background or effects layer. Read studio_film_read first. Uses the current revision and stable operationId; retry identical inputs only. No model call or credit charge.
Toggle one Film music, background or effects layer. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge.
studio_film_toggle_camerachangeddescriptionToggle whether one dialogue line’s speaker is on camera. Read studio_film_read first. Uses the current revision and stable operationId; retry identical inputs only. No model call or credit charge.
Toggle whether one dialogue line’s speaker is on camera. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge.
Showing the latest 10. Full history: list_changes · trend: quality JSON.
Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.
+dns-namespace +website +reachable +uptime-100% +updated-15d-ago -no-repository
Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed. How it is computed.
| remote | auth | reachable | uptime 30d | p50 | protocol | last ok |
|---|---|---|---|---|---|---|
https://api.brix36.com/mcp streamable-http | none | yes | 100% | 455 ms | 2026-07-28, 2025-11-25, 2025-06-18, 2025-03-26, 2024-11-05 | 2026-10-19 |
studio_account_links_readread-only | Returns verified native-app navigation and privacy/support links. No IDs selects account controls; accessible artistId/projectId adds the corresponding artist or project screens and appOnlyActions with reasons. Each ID is checked against viewer access. App-only workflows retain existing consent and spend controls. No purchase, paid work or account deletion. |
studio_account_readread-only | Read the connected customer account and its current catalogue memberships. Returns user details and up to 50 catalogues with roles. The connection determines identity; no account or workspace argument is accepted. Free. |
studio_ai_sharing_choose | Records the person's own answer to the AI-sharing question after a creation tool returned ai_processing_choice_required. allowed true means Allow and false means Not now. Consent counts only as that explicit reply; it is never inferred from the request to create and never supplied on the person's behalf. supersedes is the sequence that refusal returned (null when it was unset). Returns the stored answer; nothing is created or charged. |
studio_artist_actor_readread-only | Returns an existing artist's Film castId, identity readiness and Film creation link. castId equals artistId. No duplicate actor or artist-flag change is created. |
studio_artist_cover_set | Sets an existing passed DNA photograph as the artist cover; referenceId:null restores automatic cover selection. Requires artist permission and expectedRevision from the current profile. Returns profile and revision; creates no image. |
studio_artist_create | Creates an artist through the app service and returns profile and revision. Name is required; photos use dedicated upload/reference tools. catalogueArtistId optionally selects a manageable catalogue through an existing artist. Creation has no replay receipt; the artist list exposes persisted results after an uncertain response. No generation or charge. |
studio_artist_deletedestructive | Permanently erases an artist and their projects, storyboards, references and identity photos. Requires catalogue-owner access, the customer's deletion request and their exact current artist name as confirmation. In-flight work or payments prevent deletion. Returns artist=deleted and queued purge counts; external erasure is asynchronous. |
studio_artist_dna_readread-only | Returns identity summary, published photo pack, pending screening, approved creative facts and up to 20 recent social-scan statuses. Pending or held photos retain their status. The app link identifies consent and scan controls. No scan, language model or image generation starts. |
studio_artist_dna_write | Saves customer-authored or approved creative facts as a new published Artist DNA version, preserving facts pinned by older projects. baseVersionId identifies the current version, or null when absent. Returns versionId and facts. Facts apply at once and need no pin. Requires manager permission; no language model or credit charge. |
studio_artist_photo | Returns an artist-photo upload widget and a one-time fallback uploadUrl. File bytes go directly to Brix36. Repeated calls return the same link or report an existing photo. No charge. |
studio_artist_readread-only | Returns one artist's profile, current revision and Artist DNA summary, including genres, visual notes and identity readiness. The revision binds profile and cover edits. Read-only. |
studio_artist_ready_made_choose | Saves one Brix36 ready-made fictional portrait on an existing artist using the same service as the app. Requires artist permission and a current starId from the ready-made star list. Returns the saved reference pack and whether it was newly created; choosing the same star retains the existing pack. No image generation or credit charge. Current generation readiness is read separately from the artist list. |
studio_artist_reference_add | Adds a customer photograph to Artist DNA through screened identity-reference services. Inline input accepts canonical base64, MIME type and angle up to 512 KiB. Larger files have dedicated upload/app paths. Validates and deduplicates bytes under existing face-source/publication rules. Returns pack and created flag. No image generation. |
studio_artist_reference_deletedestructive | Remove one identity reference photo using the app service. Requires artist-level permission and the customer choosing this specific photo. Returns the updated identity pack. This can affect future likeness generation; older pinned projects keep their existing rules. |
studio_artist_reference_pack_confirmdestructive | Places the one quoted reference-pack purchase only after the customer has explicitly approved the exact credits in this chat. Takes the exact operationId, quoteToken, quotedCredits and priceVersion from the quote (and the same look inputs); an existing Brix36 approval code is accepted as confirmationToken instead. Identical retries return the first outcome and never charge twice. status still_checking keeps the original operation; an identical later call returns its outcome without a new price. |
studio_artist_reference_pack_quoteread-only | Prices the artist's missing reference views (photo_pack) or invented look (look) with the same signed price the Brix36 app shows. Charges nothing. Returns creditAmount, operationId, quoteToken, priceVersion and approvalExpiresAt. Approval of the exact credits takes place in this chat; no link is created. When nothing can be offered, says so without claiming their references are complete. Creating an artist stays free. |
studio_artists_listread-only | Lists the customer's artists alphabetically, with IDs, names, identity kind, screened photo count, face-check status and whether the real-person video model can use them. Artist IDs also identify Film cast members. Fictional characters need no photo or likeness consent; real-person artists retain the app's consent flow. Free, read-only pagination, up to 50 artists per page. |
studio_artist_update | Updates supplied artist profile fields with expectedRevision from the current profile. Returns the updated profile and revision; conflicting edits are refused. Photos, voices, identity kind and Film casting have separate operations. |
studio_artist_voice_choose | Assigns an offered ready-made voice from the voice library through a durable revision-bound operation. voiceId is an opaque catalogue ID and expectedRevision is the current voice revision, or null when no voice exists. May add a shared library voice; creates no speech and debits no credits. Film speech requires its separate quoted financial approval. Identical operation retries retain identity. |
studio_artist_voice_readread-only | Read the artist’s saved speaking voice and revision, with a private sample link when available. A recorded voice is already selected for this artist’s Film dialogue. No voice is created, no speech is generated and no credits are spent. |
studio_artist_voice_record | Creates and selects an artist's own speaking voice from an owned confirmed audio sampleAssetId, typically 30–120 seconds. App recording validation, durable operation, provider admission and daily spend guards apply. Setup cost is included in the artist's first Film quote; this call debits no customer credits. Identical operation/input retries reconcile the saved effect. Unconfigured cloning refuses before provider work. This voice is for Film dialogue, not song singing. |
studio_artist_voices_listread-only | Searches the voice library's ready-made voices. Returns offered opaque IDs, names, descriptions, selected voice, current revision and nextCursor. lib: voices use Brix36's account; acct: voices belong to connected personal accounts and are currently unavailable for Film. No speech generation or credit charge. |
studio_artist_voice_statusread-only | Returns pending, unknown, refused or succeeded for an earlier owned voice operation, with saved voice/revision on success. Unknown provider effects remain unresolved. This read does not repeat provider work or synthesize speech. |
studio_copy_readread-only | Read an owned saved customer-AI text version without generating or charging. |
studio_copy_save | Saves finished customer-authored headline/caption drafts and supporting source facts as a free text version. Identical source/revision/result inputs return the saved version. No Brix36 language-model request, publishing or credit charge. |
studio_creative_resultread-only | Returns an owned completed creative result by operationId with image.url or video.url, downloadUrl and gallery-widget data. Links are private and signed; another read refreshes expired links. The original source is retained. No processing or charge is repeated. |
studio_credits_balance_readread-only | Read the connected customer wallet. Returns productionBalance, sandboxBalance, spendableBalance, status, reviewReason and sandboxSpendingEnabled from the app ledger. A review_required wallet cannot spend. Free; purchases remain in the Brix36 app. |
studio_credit_usage_listread-only | Lists the connected customer's purchases, spends, refunds and adjustments newest first. Returns signed amounts, source, time, accessible projectTitle and nextCursor. Pagination preserves the cursor value. No wallet changes. |
studio_film_add_linedestructive | Add one finished dialogue line to a scene and return its lineId. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_assign_actordestructive | Assign one available artist to one saved dialogue line. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_direction_setdestructive | Replace one scene’s direction and reference pictures. A null sceneIndex with empty text binds whole-Film references. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_edit_linedestructive | Replace the text of one saved dialogue line. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_readread-only | Returns saved Film scenes, dialogue, cast, voice assignments, planning state and savedPlan.revision. Add include:['languages'] for supportedLanguages and nativeSpeechLanguages (the video model's own speech), and include:['artists'] for every artist who can join the cast; leave include out for the project view. nativeSpeechEngines holds model-specific evidence. Saved per-scene voice choices determine routing; absent saved voices use native speech. Audio-off prices are unavailable for speaking scenes. operationId reads prior plan progress without starting or settling work. |
studio_film_remove_linedestructive | Remove one saved dialogue line. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_scenes_write | Saves finished Film scenes, visualWorld, camera directions and dialogue against the current revision. Scenes form 30-second cards, with at most six spoken lines per card and five minutes per Film. Identical operationId/input retries return the saved result. visualWorld is required for quoting. No Brix36 planning model, media generation or credit charge. |
studio_film_script_write | Imports finished screenplay-style Film text, with SCENE headings, NAME: dialogue and optional visualWorld. Preserves scenes, dialogue and speaker names, caps Film at five minutes and reports uncast speakers. Identical retries return the same plan. A visualWorld is required for quoting. No Brix36 planning model or credit charge. |
studio_film_set_durationdestructive | Set one scene’s duration, normalized to 4–30 seconds; later scenes move to follow it. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_set_lip_syncdestructive | Set the lip-sync choice for one saved scene. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_set_speech_voicedestructive | Set the saved-character or native speech voice choice for one character in one scene. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_toggle_cameradestructive | Toggle whether one dialogue line’s speaker is on camera. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_film_toggle_sounddestructive | Toggle one Film music, background or effects layer. Uses expectedRevision, the current saved Film plan revision, and a stable operationId; identical operationId/input retries reconcile the same edit. No model call or credit charge. |
studio_generation_confirmdestructive | Spends Brix36 credits and starts the quoted work under the customer's financial approval. In chat mode, stage:pictures records one approval for pictures and video using the original operationId, quoteToken, priceVersion, quotedCredits and revisionId. Insufficient balance returns balance, shortfall and creditsInformationUrl without a checkout. Stage:video uses the current picture revision and the same financial approval, subject to the existing readiness gates. Identical stage and argument retries reconcile the same operation. New authorizations retain pause/cap, OAuth studio.spend, account and exact-quote checks. The response waits up to 45 seconds for the started stage and includes current progress, a finished video link when available, or the free status-call arguments for work still rendering. Existing page-approved operations and confirmationToken codes remain supported. |
studio_generation_quoteread-only | Prices missing storyboard pictures plus the exact video price for one run without spending credits. Returns totalCredits, totalGbpInclVat, balance, shortfall, signed quoteToken, operationId, revisionId and priceVersion. engineRecommendations contains server-ranked choices, current selections and quoted alternatives; section engine selection returns a revised quote. Film quotes include per-character, per-scene speech, motion and lip-sync routes, or historical_unavailable when prior evidence is absent. Chat-mode confirmation uses these exact quote fields. The saved financial approval covers pictures and video; requested picture changes have separate quotes. No checkout or approval-page URL is created. Existing app approvals and issued codes remain supported. |
studio_generation_statusread-only | Waits up to 45 seconds for an active rendering stage to finish, then returns its current state and continuation data. Reads a saved whole-job or picture-only operation by its original projectId and operationId. Returns approval and generation state, next-step data, balance or shortfall where applicable, and any fresh unapproved quote. Whole-job states include making_pictures, pictures_ready, video_waiting_for_payment, starting_video, making_video and failed. Picture-only approval covers only its quoted pictures. Video retains the original financial approval and applicable readiness gates; likeness not_required needs no likeness verdict. This free read does not advance stages or charge credits. |
studio_generation_waitread-only | Waits up to 45 seconds for active picture generation or the finished video. projectId identifies the video; optional operationId identifies its original picture/video approval. A pending response is a nonterminal timeout containing the exact nextCall for another bounded wait. A completed video response contains video.url. Free and read-only; no generation, approval, quote, payment or credit change. |
studio_get_skillread-only | Returns one Brix36 creative skill, style card or writing guide by catalogue ID, including full body, version, hash and resource URI. Read-only. |
studio_image_confirmdestructive | Spends Brix36 credits and starts one picture under the customer's financial approval of the exact quoted draftId and quoteId. Identical retries return the same operation without another charge. |
studio_image_quoteread-only | Prices one standalone picture from a text prompt and optional owned uploads, using the same pricing as the Brix36 create box. Returns the exact credits and price with draftId and quoteId. Nothing is generated or charged. |
studio_image_resultread-only | Returns the owned picture for an approved quoteId with image.url, downloadUrl and gallery-widget data once ready, or its current progress. Links are private and signed; another read refreshes them. No processing or charge is repeated. |
studio_list_skillsread-only | Searches Brix36 creative skills and style cards by keyword, medium or family. Returns summaries rather than full bodies, plus resource URIs and nextCursor. No Brix36 language-model call. |
studio_media_confirmread-only | Returns pending or accepted attachment status and assetId for the exact owned uploadToken. Reconciles the existing upload without creating a second attachment, generating media or spending credits. |
studio_media_import_url | Import a public HTTPS song, artist photo or reference into the same private attachment pipeline as an upload. projectId is optional in the schema but required to attach media: omission returns guidance to select a project, never another user project. Refuses private/loopback destinations, unsafe redirects, DNS rebinding, invalid file types, oversized downloads and slow sources. Song/photo replacement rules remain unchanged. Identical reference bytes in the same project replay the existing asset. Returns assetId and status; no provider or language model is called. |
studio_media_upload_url | Creates an HTTPS POST upload capability bound to the user, project, role, exact file metadata and SHA-256, expiring in 15 minutes. Returns uploadUrl, uploadToken, method and expiresAt. The upload accepts raw bytes with the declared MIME Content-Type; attachment status is confirmed separately. No file bytes pass through MCP and no generation credits are spent. |
studio_media_upload_widgetread-only | Opens a file picker for a song, artist photo or image/video reference on the specified project. The customer selects the file; bytes go directly to Brix36. No paid work. |
studio_model_catalogueread-only | Returns the current Brix36 model catalogue, including profiles, settings, media limits, availability reasons, recommended choices and qualified starting prices in credits and GBP minor units including VAT. Starting prices are distinct from generation quotes. Pagination uses nextCursor with the same kind. No generation or charge. |
studio_motion_catalogueread-only | List available Motion Graphics templates (lyric_video, audio_visualiser, living_cover, kinetic_type, club_tunnel, organic_landscape), 7 genre looks (pop, hiphop_drill, rnb, house_edm, rock, country_acoustic, afrobeats), and all 24 popular motion styles with preview video URLs and input requirements. Read-only discovery. |
studio_motion_confirmdestructive | Starts the exact paid server render after the customer explicitly approves its quoted price. Chat approval returns operationId, quoteToken, revisionId, inputDigest and quotedCredits unchanged from the quote. Existing Brix36 confirmationToken codes remain supported as a separate flow. The server enforces spending permission, account policy and exact quote bindings. Identical retries reconcile the original operation. Annotations do not attest to consent or guarantee a host confirmation screen. |
studio_motion_create | Saves a ready server-rendered Motion scene and returns its exact paid-render quote plus an optional free first-frame preview. Supports lyric video, beat-reactive visualiser, kinetic titles, photo-motion poster, promo cards, the 6 music/lyric/visualiser templates (lyric_video, audio_visualiser, living_cover, kinetic_type, club_tunnel, organic_landscape) across 7 genre looks (pop, hiphop_drill, rnb, house_edm, rock, country_acoustic, afrobeats), and all 24 popular motion styles (styleId). Uses owned audio and picture asset IDs; saved song timing supplies lyric cues and the saved scene can be edited. No AI video model or paid render is called. requestId identifies the saved document across retries. Each returned quote has its own identity; rendering requires explicit customer approval bound to that quote. |
studio_motion_previewread-only | Generate a free bounded first-frame image of this exact saved revision, maximum 640px long side. No LLM, render debit or full free movie. Preview is not an approval or a priced render. |
studio_motion_quoteread-only | Returns the server price in credits and GBP including VAT, bound to the exact revision, output, source and runtime. A paid render requires one explicit customer approval of this exact price. Edits invalidate prior quotes. No debit or render. |
studio_motion_readread-only | Read the exact authored scene and revision, including editable lyric cue timing, text and owned media layers. |
studio_motion_restore | Restores a saved scene as a new revision with optimistic version checking. Original versions and media remain intact. Prior quotes become stale; further paid rendering requires a quote for the restored revision. |
studio_motion_resultread-only | Read the finished owned asset with video.url and authenticated playback/download links. Source media remain preserved. Pending work is not a completed result. |
studio_motion_statusread-only | Read the original durable operation after pending, timeout or unknown outcome. Successful work includes playback and download links. Reconcile means the existing effect is being checked; no replacement render is started. |
studio_motion_update | Saves typed MotionSceneV1 with exact words, cue frame ranges and owned asset IDs. Accepts typed layers and approved fonts; HTML, JS, CSS and URLs are not scene inputs. First save creates an opaque documentId with expectedRevision 0; edits bind the current expectedRevision. Conflicts preserve the newer saved version. No Brix36 language model is called. |
studio_motion_use | Add the exact finished Motion result to project media through normal attachment screening, preserving original assets, movie orders and expected project revision. This does not replace or regenerate a paid movie. |
studio_motion_waitread-only | Waits up to 40 seconds for the original Motion operation, then returns its finished video.url and downloadUrl or a nonterminal pending response with an exact nextCall. Each status read checks current access. No generation, approval or credit change. |
studio_picture_use_upload | Sets an owned uploaded photograph as a shot's original picture, bound to attachment contentHash and confirmed rights. Creates a canonical revision and retains earlier versions. Identical argument retries reconcile the same operation. No generation or credit charge. |
studio_project_artist_presence_setdestructive | Set featured or not_featured for this project. Featured requires pinned DNA; the existing directed-storyboard rules still apply. No generation or charge. |
studio_project_deletedestructive | Archive one video or Film using the app project deletion service after the customer requests deletion. Requires editor permission. In-flight orders, paid work awaiting consumption and open/review-required checkouts refuse deletion. Returns deleted=true. This is separate from permanently erasing an artist or account. |
studio_project_dna_pindestructive | Pin this artist’s currently published DNA to this project once, within its existing pin window. Read project state after a lost response before retrying. No generation or charge. |
studio_project_palette_write | Save 0–5 unique uppercase #RRGGBB colours as exact project creative direction. An empty list clears the selection. Existing work remains in history; stale revision and active work refuse safely. |
studio_project_rename | Renames a video or Film using expectedRevision, title and optional songTitle. Returns updated project and revision. No generation or charge. |
studio_projects_listread-only | Lists the customer's videos newest first, optionally filtered by artistId. Returns projectId, title, artist, status, picture, last-change time and nextCursor. Free and read-only, up to 50 projects per page. |
studio_ready_made_stars_listread-only | Lists the same free fictional character portraits offered by the Brix36 app, with each current star ID, name, description and versioned preview image. Read-only; creates no artist, image, generation or charge. |
studio_reference_add | Stores and screens an image/video reference through the app attachment service. Inline files are limited to 512 KiB; larger files have an app upload-link path. Returns attachment and signed preview. scope:project applies to the project; scope:shot supports later shot binding. The reference list exposes persisted results after uncertain replies. No language model or generation-credit charge. |
studio_reference_deletedestructive | Remove an existing image/video reference attachment from a project using the app service. Requires editor permission and the customer choosing this reference. The app removes its dependent notes and applies existing reference binding rules. Returns deleted=true; it does not delete another project reference or a song. |
studio_reference_note_add | Save one concrete creative instruction about a customer reference already on this project. The connected assistant does the interpretation with the customer; Brix36 validates and stores the finished sentence through the existing app fact service. Returns an approved user-authored fact. Duplicate category/sentence on the same attachment is reused. Free; no server language model is called. |
studio_reference_notes_listread-only | Returns a paginated list of existing approved, proposed or rejected reference notes for a project, optionally filtered by attachment. No server interpretation starts. |
studio_reference_note_updatedestructive | Approve, reject or rewrite one reference note using the app service. Rewriting records the sentence as user-authored; omitted fields stay unchanged. Returns the updated fact. Requires permission to edit its project and the customer choosing the change. No server interpretation or paid work starts. |
studio_references_listread-only | Browse customer image/video references across their accessible projects, or filter one project. Returns up to 50 references with projectId, attachmentId, signed contentUrl, scope, analysis status and a small preview of existing facts, plus nextCursor. Shot references affect only bound shots. Nothing is analysed by this tool. |
studio_reference_upload_linkread-only | Returns the app reference-upload screen for an editable project, including for files over the 512 KiB inline limit. Uploaded results appear in the project's reference list. No upload or generation occurs in this read. |
studio_release_create | Creates an album, EP or single collection with title, type, year and artwork preset. Returns release summary and app project-creation links. Creation has no replay receipt; the artist's release list exposes saved title/type/year after an uncertain response. No language model, generation or charge. |
studio_releases_listread-only | Lists an artist's album, EP and single collections with projectIds, status, artwork, year, verified app project-creation links and nextCursor. Free and read-only. MCP project creation currently creates standalone releases; collection project creation uses the app. |
studio_section_enginesread-only | Returns engine options for one section of an existing generation quote identified by costEstimateId or quoteToken. Film sectionIndex is the saved zero-based scene.index. Includes availability, requirements, selection, server Recommended marker, allocated_final_total section share, availableProfiles and finalVideoCredits. Option selection_total is a whole alternative video quote total, excluding pictures. No selection change or spending. |
studio_section_engine_select | Selects an available modelName or generation profile for one quoted section and returns a revised signed quote with costEstimateId. Film uses saved scene.index. modelName:null restores the server-ranked Recommended choice; omitted modelName preserves explicit selections. profile {aspect,resolution} is constrained to availableProfiles. Other sections retain their choices; outputs fit the project canvas without cropping. The revised quote has its own price and approval binding. operationId identifies identical selection retries. Film speech data reports the actual per-scene routes or historical_unavailable. No credits are spent and no generation starts. |
studio_sections_readread-only | Returns the project's 30-second storyboard sections, including indexes, times, names, inclusion, pictures, action, Performance/Story type and lyric line IDs with lip-sync settings. These indexes and IDs identify inputs to section-edit tools. Free and read-only. |
studio_sections_set_line_sync | Sets lip-sync for one existing lyric line identified by section index and stable line ID. sync:true means the artist mouths the line. Defaults are true for the artist's own words and false for guest/ad-lib lines. No generation or charge. |
studio_sections_update | Updates included, type and/or what for one existing 30-second storyboard section. Omitted fields remain unchanged. Timing, names and pictures are outside this operation. No generation or charge. |
studio_shot_video_references_readread-only | Read the actual owned driving-video references for a shot. These are distinct from extracted reference pictures. No paid work. |
studio_shot_video_references_write | Binds up to three owned uploaded 1–15 second original clips as shot motion/composition references. Measures bytes and creates one canonical revision. Stricter model limits or unsupported driving-video inputs refuse before payment. Clips remain video references rather than sampled pictures. Empty attachmentIds clears the selection; identical retries reconcile the same write. |
studio_song_attach | Returns a song upload widget and a one-time fallback uploadUrl for an existing project. File bytes go directly to Brix36. Repeated calls return the same fallback link or the existing song attachment. No charge. |
studio_song_reuse | Attaches existing customer audio from an accessible source project to an empty target through the durable upload/analysis path. sourceAttachmentId identifies the saved song attachment. Returns attachment/asset/job IDs or already_attached for identical audio. Repeats return the existing result; different current audio refuses replacement. No arbitrary URL fetch, generated audio or language-model call. |
studio_songs_listread-only | List one artist songs using the app canonical grouping, including their available project versions. Returns a bounded songs page and nextCursor. Each song projectId identifies the project holding that song version. Free; this does not upload or generate audio. |
studio_song_statusread-only | Read the currently attached song and its technical analysis for a project. Returns the song attachment, latest upload job status, analysis ID, measured duration, tempo, sections and existing themes when available. Absent analysis stays pending. Free; this never starts transcription, reasoning or a storyboard. |
studio_song_understanding_write | Saves finished song understanding against the current analysisId, assetId and understandingRevision. Accepts lyrics/description, language, themes and required words/lines arrays in the timed-transcription shape; empty arrays represent unavailable genuine timing. Source text alone supplies no lip-sync timing or voice ownership. voiceAnswer represents the customer's all_me, other_voices or skipped decision. Preserves measured audio facts and returns canonical understanding with a new revision. Stale input is refused. No transcription/planning model or credit charge. |
studio_storyboard_approve | Records the customer's storyboard approval for the exact current revisionId using the app's approval service. Changed revisions are refused. Directed-board and existing purchase guards remain enforced. No credits are spent and no media is generated. |
studio_storyboard_contextread-only | Returns current song analysis, brief, approved Artist DNA, references, exact shot timings and storyboard for a Music project. Includes audioAnalysisId, expectedRevisionId, fixed skeleton, approved evidence and submission schema for direction and visualWorld. No planning model or other work starts. |
studio_storyboard_edit_prompt | Saves finished prompt text and existing uploaded references as an immutable storyboard revision. Whole/section scope writes shared sectionDirection; shot/still scope writes frame description and optional motion/camera. References replace targeted bindings; [] clears and omission inherits. Requires the current revision. No pictures, video, planning model or credit charge. |
studio_storyboard_likeness_recorddestructive | Record the customer’s confirmed or rejected likeness verdict for one exact saved picture. Requires the revision, shot and still asset the customer reviewed. Changes no picture and spends no credits. |
studio_storyboard_look_acceptdestructive | Record the customer’s acceptance of the complete saved key-look picture set. Supply the exact revision and shot/asset pairs reviewed. Changes no picture and spends no credits. |
studio_storyboard_picture_changes_preparedestructive | Prices requested picture changes by original projectId/operationId, a new requestId and selected stableShotIds. New approvals produce separate picture-only quotes containing operationId, quoteToken, revisionId, modelId, mode, priceVersion and totalCredits. Identical requestId retries return the original quotes. Historical approvals can complete their previously approved paid picture changes. No charge for quoting. |
studio_storyboard_prepare | Validates and saves visualWorld for an existing Music or Film storyboard at expectedRevisionId. Preserves scenes, dialogue, timing, stable shot IDs, cast, identity choices and references. Returns ready synchronously for a valid world, or needs_visual_world with saved-scene context and schema when absent. An already-valid saved world returns ready without a new revision. No planning model, queue, generation provider or credit use. |
studio_storyboard_previewread-only | Returns the stored storyboard as 30-second section cards with lines and numbered shots. Each shot includes stableShotId and a signed stillUrl valid for about a day, or null if no picture exists. Supported hosts also receive an interactive preview. Free and read-only. |
studio_storyboard_readinessread-only | Returns readyForVideo, shotsMissingStills, public likenessGate, look-test state and reviewStills for the exact project and revision. Each review binding identifies a saved stableShotId/stillAssetId. A likenessGate of not_required needs no likeness verdict; an absent look test needs no look acceptance. Required likeness verdicts and acceptance of the saved look set are recorded separately. Video uses the original approved price and the existing readiness gates. No state change, provider call or charge. |
studio_storyboard_redo_requestread-only | Validates a widget-selected stableShotId against the current visible storyboard and returns a request to price that picture change. No approval, generation or spending. |
studio_storyboard_stills_confirmdestructive | Spends Brix36 credits for exactly the approved quoted pictures. Chat mode requires the quote's unchanged projectId, revisionId, mode, operationId, quoteToken, modelId, priceVersion and quotedCredits. Insufficient balance returns balance, shortfall and creditsInformationUrl without a checkout. Durable purchase/job reconciliation makes identical retries safe, including after expiry. Legacy confirmationToken remains supported. No video authorization or planning-model call. |
studio_storyboard_stills_quoteread-only | Prices selected storyboard pictures without spending. Supports missing pictures or new_version for exactly one existing picture, bound to the current revision, quality and stableShotIds. lookTest:true selects the saved bounded key-look set. Returns a signed picture-only quote and exact confirmation fields. Missing operationId is derived by the server; new_version uses a stable op_ ID. Chat approval and existing issued confirmation codes are supported. Short balance returns credit information only. This quote covers no video or planning model. |
studio_storyboard_submit | Validates and saves a finished Music storyboard with one DirectedShot per fixed shot order, exact expectedRevisionId/audioAnalysisId, optional treatment and visualWorld. Creates an immutable revision while preserving server-owned timing. visualWorld is required for quoting. No model, queue, provider or credit use. |
studio_storyboard_write | Updates the supplied existing 30-second storyboard sections with what, type and included fields. The server owns section indexes, count, names, timing and pictures. All supplied edits succeed together or the prior board is restored with a refusal identifying the failed section. No generation or credit charge. |
studio_video_exportread-only | Returns signed download links for existing finished Music video or Film results and selected range files, including results beside unfinished edits. No public publishing, render, payment or extra export charge. |
studio_video_plandestructive | Creates or updates a film or music-video project, saves its script or storyboard, and returns a price quote without charging. Film input contains the finished script, language, cast and visualWorld. Music input contains sections, visualWorld and optional songUnderstanding. start identifies a new project; projectId identifies an existing one. A music project without a song returns state needs_song and its projectId; planning can continue on that project after its song upload. The result contains the project, saved plan and quote. Identical operationId and input retries return the same project and quote. Rendering is separate and requires the customer's approval of the quoted price. |
studio_video_previewread-only | Returns current edit status and signed playback links for an existing finished master or selected range files, including finished results beside unfinished edits. Supported hosts receive players. No generation or credit charge. |
studio_video_start | Creates a Music or Film project in the customer's catalogue. Accepts an existing artist name or a new fictional name with createArtist:true. Fictional characters require no photo or likeness consent; real-person creation retains the app consent flow. Music requires songTitle; Film requires title or songTitle and no song. Identical retries return the same project. Returns project identifiers for song upload or Film script import. No generation or charge. |
studio_video_statusread-only | Waits up to 45 seconds while rendering, then returns draft, storyboard_ready, making, ready or needs_attention state, section progress counts, the studio's explanation and video.url when finished, with selected range labels, extent and original file links. Excluded sections are counted separately from failures. Provider fallback remains within the quoted price. No new work or charge. |
Schemas: list_tools.
{
"name": "com.brix36/brix36",
"_meta": {
"io.modelcontextprotocol.registry/publisher-provided": {
"com.brix36": {
"supportUrl": "https://brix36.com/support",
"serverCardUrl": "https://api.brix36.com/.well-known/mcp/server-card.json",
"authentication": "OAuth 2.1 with PKCE",
"privacyPolicyUrl": "https://brix36.com/privacy",
"termsOfServiceUrl": "https://brix36.com/terms"
}
}
},
"icons": [
{
"src": "https://api.brix36.com/icon.png",
"mimeType": "image/png"
}
],
"title": "Brix36",
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"remotes": [
{
"url": "https://api.brix36.com/mcp",
"type": "streamable-http"
}
],
"version": "0.2.0",
"websiteUrl": "https://brix36.com/connect",
"description": "Films and music videos. You're the star. No subscription."
}