{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"smithery_acedatacloud-mcp-mcp-luma","slug":"acedatacloud-mcp-mcp-luma","name":"mcp-luma","type":"mcp","url":"https://smithery.ai/servers/acedatacloud-mcp/mcp-luma","page_url":"https://unfragile.ai/acedatacloud-mcp-mcp-luma","categories":["mcp-servers"],"tags":["mcp","model-context-protocol","smithery:acedatacloud-mcp/mcp-luma"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"smithery_acedatacloud-mcp-mcp-luma__cap_0","uri":"capability://tool.use.integration.luma.ai.video.generation.via.mcp.protocol","name":"luma ai video generation via mcp protocol","description":"Exposes Luma AI's video generation capabilities through the Model Context Protocol, allowing Claude and other MCP-compatible clients to invoke video creation without direct API integration. Implements MCP's resource and tool abstractions to translate high-level generation requests into Luma API calls, handling authentication, polling for async job completion, and streaming results back through the MCP transport layer.","intents":["Generate videos from text prompts within Claude conversations without leaving the chat interface","Integrate Luma video generation into multi-step AI workflows that combine reasoning, planning, and media creation","Build AI agents that can autonomously create video content as part of task execution","Access Luma's video models (Dream Machine, Photon) programmatically through a standardized protocol"],"best_for":["AI application developers building Claude-integrated video workflows","Teams deploying MCP servers in enterprise Claude deployments","Builders creating multi-modal AI agents that need video generation capabilities"],"limitations":["Async job model means video generation requests don't return immediately — requires polling or callback handling","Rate limiting and quota constraints from Luma API tier apply directly to MCP server throughput","No built-in caching of generated videos — each request triggers a new generation even for identical prompts","MCP transport adds serialization overhead compared to direct REST API calls"],"requires":["Luma AI API key with video generation quota","MCP-compatible client (Claude Desktop, custom MCP host, etc.)","Network connectivity to Luma API endpoints","Python 3.8+ or Node.js 16+ depending on server implementation"],"input_types":["text prompts (natural language video descriptions)","structured generation parameters (duration, aspect ratio, style)","optional seed values for reproducibility"],"output_types":["video file URLs (MP4 or WebM format)","job status objects with generation metadata","structured generation results with timing and model information"],"categories":["tool-use-integration","image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_acedatacloud-mcp-mcp-luma__cap_1","uri":"capability://automation.workflow.async.video.generation.job.orchestration.with.polling","name":"async video generation job orchestration with polling","description":"Manages the asynchronous lifecycle of Luma video generation requests by implementing a polling-based job tracker that monitors generation status, handles retries on transient failures, and surfaces job metadata (progress, estimated completion time, error states) back to the MCP client. Abstracts Luma's job ID-based tracking into a stateful resource model compatible with MCP's resource protocol.","intents":["Monitor long-running video generation tasks without blocking the MCP client","Retrieve generation status and progress updates for videos in-flight","Handle generation failures gracefully with retry logic and error reporting","Correlate multiple video generation requests and track their completion states"],"best_for":["Long-running AI workflows that generate multiple videos sequentially or in parallel","Applications requiring visibility into video generation progress and ETA","Teams building resilient agents that need to handle generation timeouts and retries"],"limitations":["Polling introduces latency — status updates are not real-time, typically 1-5 second intervals","No webhook support means the MCP server must actively poll Luma API, consuming API quota","Job state is not persisted across server restarts — in-flight generations are lost if the MCP server crashes","Polling frequency must be tuned to balance latency vs API rate limit consumption"],"requires":["Luma API key with sufficient quota for polling requests","MCP client capable of handling async resource updates","Timeout configuration for maximum polling duration (typically 5-30 minutes)"],"input_types":["job IDs from previous generation requests","polling interval configuration (seconds)","timeout thresholds"],"output_types":["job status objects (pending, processing, completed, failed)","progress metadata (estimated time remaining, processing stage)","error details and retry recommendations"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_acedatacloud-mcp-mcp-luma__cap_2","uri":"capability://tool.use.integration.mcp.resource.exposure.for.generated.video.artifacts","name":"mcp resource exposure for generated video artifacts","description":"Exposes generated videos and their metadata as MCP resources, allowing Claude and other MCP clients to reference, retrieve, and reason about video generation outputs within the protocol's resource model. Implements MCP's resource URI scheme to make videos queryable and linkable, with support for metadata annotations (generation parameters, model used, creation timestamp).","intents":["Reference generated videos in Claude conversations by resource URI instead of raw URLs","Retrieve metadata about generated videos (model, parameters, generation time) for analysis or logging","Build multi-step workflows where video generation output feeds into downstream tasks","Enable Claude to reason about video generation history and outputs within a conversation"],"best_for":["Claude users building video-centric workflows that need artifact tracking","Teams deploying MCP servers that need to expose media artifacts as first-class resources","Applications requiring audit trails and metadata for generated videos"],"limitations":["Resource URIs are server-scoped — videos generated by one MCP server instance are not directly accessible from another","No built-in versioning or branching of video artifacts — each generation is a separate resource","Metadata is limited to what Luma API returns — custom annotations require server-side storage","Resource cleanup and garbage collection must be manually configured to avoid unbounded storage"],"requires":["MCP client that supports resource protocol (Claude Desktop 0.4+)","Server-side storage or caching for video metadata","URI scheme configuration for resource identification"],"input_types":["video generation job IDs","custom metadata annotations (optional)"],"output_types":["MCP resource URIs (e.g., mcp://luma/video/{job_id})","structured metadata objects with generation parameters and timestamps","video file URLs and format information"],"categories":["tool-use-integration","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_acedatacloud-mcp-mcp-luma__cap_3","uri":"capability://tool.use.integration.tool.based.video.generation.parameter.schema.validation","name":"tool-based video generation parameter schema validation","description":"Exposes video generation as an MCP tool with a strict JSON schema that validates input parameters (prompt, duration, aspect ratio, style, seed) before sending to Luma API. Uses schema-based validation to catch invalid parameter combinations early, provide helpful error messages, and ensure generated requests conform to Luma's API constraints. Implements parameter normalization (e.g., aspect ratio formatting, duration clamping) to handle client variations.","intents":["Validate video generation parameters before API calls to avoid wasted quota on invalid requests","Provide Claude with clear parameter constraints and valid options for video generation","Normalize parameter formats across different client implementations","Generate helpful error messages when users provide invalid or conflicting parameters"],"best_for":["Teams deploying MCP servers that need to enforce API constraints at the protocol level","Applications where invalid API calls are costly (quota-limited APIs like Luma)","Builders creating user-facing interfaces that need to guide parameter selection"],"limitations":["Schema validation is static — cannot adapt to Luma API changes without server updates","Parameter normalization may silently alter user intent (e.g., clamping duration) without explicit feedback","Schema complexity grows with Luma API feature additions — maintenance burden increases","No support for conditional parameter validation (e.g., some parameters only valid with specific models)"],"requires":["JSON Schema definition for Luma video generation parameters","MCP tool protocol support in client","Parameter validation library (e.g., jsonschema, zod)"],"input_types":["text prompt (string, 1-1000 characters)","duration (integer, 5-120 seconds)","aspect ratio (enum: 16:9, 9:16, 1:1)","style (enum: cinematic, realistic, artistic, etc.)","seed (optional integer for reproducibility)"],"output_types":["validation success/failure status","normalized parameter object","error messages with constraint details"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_acedatacloud-mcp-mcp-luma__cap_4","uri":"capability://safety.moderation.authentication.and.credential.management.for.luma.api","name":"authentication and credential management for luma api","description":"Manages Luma API authentication by securely storing and injecting API keys into requests, supporting multiple credential sources (environment variables, configuration files, credential stores). Implements credential refresh logic for token-based auth if Luma supports it, and provides error handling for authentication failures with clear messaging. Abstracts credential management from the MCP client, keeping secrets server-side.","intents":["Securely store Luma API credentials without exposing them to MCP clients","Support multiple credential sources and fallback mechanisms","Handle authentication errors gracefully with clear error messages","Rotate or refresh credentials without restarting the MCP server"],"best_for":["Enterprise deployments where credentials must be managed server-side","Teams deploying MCP servers in shared environments","Applications requiring credential rotation or multi-tenant credential isolation"],"limitations":["Credentials must be provisioned to the MCP server at startup — no dynamic credential injection from clients","No built-in credential rotation — requires external orchestration (e.g., Kubernetes secrets)","Single credential per server instance — no per-user quota tracking or isolation","Credential leaks in server logs or error messages are possible if not carefully handled"],"requires":["Luma API key (from Luma AI account)","Secure credential storage mechanism (environment variables, secrets manager, etc.)","MCP server process with appropriate file/environment access"],"input_types":["API key (string, typically 32-64 characters)","credential source configuration (env var name, file path, etc.)"],"output_types":["authentication success/failure status","error messages (without exposing credentials)"],"categories":["safety-moderation","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":24,"verified":false,"data_access_risk":"high","permissions":["Luma AI API key with video generation quota","MCP-compatible client (Claude Desktop, custom MCP host, etc.)","Network connectivity to Luma API endpoints","Python 3.8+ or Node.js 16+ depending on server implementation","Luma API key with sufficient quota for polling requests","MCP client capable of handling async resource updates","Timeout configuration for maximum polling duration (typically 5-30 minutes)","MCP client that supports resource protocol (Claude Desktop 0.4+)","Server-side storage or caching for video metadata","URI scheme configuration for resource identification"],"failure_modes":["Async job model means video generation requests don't return immediately — requires polling or callback handling","Rate limiting and quota constraints from Luma API tier apply directly to MCP server throughput","No built-in caching of generated videos — each request triggers a new generation even for identical prompts","MCP transport adds serialization overhead compared to direct REST API calls","Polling introduces latency — status updates are not real-time, typically 1-5 second intervals","No webhook support means the MCP server must actively poll Luma API, consuming API quota","Job state is not persisted across server restarts — in-flight generations are lost if the MCP server crashes","Polling frequency must be tuned to balance latency vs API rate limit consumption","Resource URIs are server-scoped — videos generated by one MCP server instance are not directly accessible from another","No built-in versioning or branching of video artifacts — each generation is a separate resource","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.2,"ecosystem":0.38999999999999996,"match_graph":0.25,"freshness":0.5,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.15,"match_graph":0.23,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:25.061Z","last_scraped_at":"2026-05-03T15:18:42.145Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=acedatacloud-mcp-mcp-luma","compare_url":"https://unfragile.ai/compare?artifact=acedatacloud-mcp-mcp-luma"}},"signature":"FYjBU1v+yDmvHYKqrXFfvfIHXcVaN9TPwPpE/PXSnEgztzKOMI88HHNrdWBGIj0TB12BIFN7A5GyOeHajjUZCg==","signedAt":"2026-06-22T17:30:58.194Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/acedatacloud-mcp-mcp-luma","artifact":"https://unfragile.ai/acedatacloud-mcp-mcp-luma","verify":"https://unfragile.ai/api/v1/verify?slug=acedatacloud-mcp-mcp-luma","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}