{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-langgenius--dify","slug":"langgenius--dify","name":"dify","type":"platform","url":"https://dify.ai","page_url":"https://unfragile.ai/langgenius--dify","categories":["automation","deployment-infra"],"tags":["agent","agentic-ai","agentic-framework","agentic-workflow","ai","automation","gemini","genai","gpt","gpt-4","llm","low-code","mcp","nextjs","no-code","openai","orchestration","python","rag","workflow"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-langgenius--dify__cap_0","uri":"capability://tool.use.integration.multi.provider.llm.model.invocation.with.quota.management","name":"multi-provider llm model invocation with quota management","description":"Dify implements a Provider and Model Architecture that abstracts multiple LLM providers (OpenAI, Anthropic, Gemini, etc.) through a unified invocation pipeline. The system uses a quota management layer with credit pools to track and limit API consumption per tenant, enforcing rate limits and cost controls at the model invocation level before requests reach external APIs. This architecture enables seamless provider switching and cost governance across multi-tenant deployments.","intents":["Switch between LLM providers without changing workflow definitions","Track and enforce API usage quotas per workspace or user","Implement cost controls and prevent runaway API spending","Support multiple model families in a single application"],"best_for":["Teams building multi-tenant SaaS platforms with LLM backends","Organizations needing fine-grained cost control and usage tracking","Developers avoiding vendor lock-in by supporting multiple LLM providers"],"limitations":["Quota enforcement is soft-limit only — does not prevent requests already in-flight when quota exhausted","Provider-specific features (e.g., vision, function calling) require explicit adapter implementation per provider","No built-in fallback mechanism if primary provider quota is exhausted"],"requires":["API keys for at least one supported LLM provider (OpenAI, Anthropic, Gemini, etc.)","PostgreSQL or compatible database for quota state persistence","Python 3.9+ for backend API service"],"input_types":["text prompts","structured model parameters (temperature, max_tokens, etc.)","provider-specific configuration objects"],"output_types":["text completions","structured JSON responses","streaming token sequences","usage metrics (tokens, cost)"],"categories":["tool-use-integration","llm-provider-abstraction"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_1","uri":"capability://planning.reasoning.workflow.engine.with.node.based.dag.execution.and.pause.resume","name":"workflow engine with node-based dag execution and pause-resume","description":"Dify's Workflow Engine uses a Directed Acyclic Graph (DAG) execution model where workflows are composed of typed nodes (LLM, HTTP, Code, Knowledge Retrieval, Human Input) connected by edges. The engine executes nodes sequentially or in parallel based on dependencies, with a pause-resume mechanism that allows Human Input nodes to block execution and wait for external input before continuing. Node Factory and Dependency Injection patterns enable dynamic node instantiation and testing via mock systems.","intents":["Define complex multi-step AI workflows without coding","Pause workflows at human decision points and resume with user input","Execute HTTP requests, code snippets, and knowledge retrieval in orchestrated sequences","Test workflows with mock data before production deployment"],"best_for":["Non-technical users building AI automation workflows","Teams implementing approval workflows with human-in-the-loop steps","Developers prototyping complex agent behaviors before coding"],"limitations":["DAG model does not support cycles or loops — workflows must be acyclic","Pause-resume state is stored in-memory by default; requires external persistence for multi-instance deployments","Node execution is sequential within a single workflow instance; no true parallel execution across nodes","Debugging paused workflows requires manual state inspection; no built-in debugger"],"requires":["PostgreSQL for workflow definition and execution state storage","Redis for distributed task queue (Celery) if using background execution","Python 3.9+ for backend workflow engine"],"input_types":["workflow JSON definitions","node configuration objects","runtime variables and context","user input for Human Input nodes"],"output_types":["workflow execution logs","node output values","final workflow result","execution trace with timing"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_10","uri":"capability://automation.workflow.application.deployment.with.docker.and.multi.environment.configuration","name":"application deployment with docker and multi-environment configuration","description":"Dify provides Docker Build Process with Multi-Stage Images for containerized deployment, supporting both API and frontend services. The system uses Environment Configuration and Runtime Modes to manage settings across development, staging, and production environments. Docker Compose Stack orchestrates the full application stack (API, frontend, PostgreSQL, Redis, vector database) for local development and testing, while production deployments use Kubernetes or managed container services.","intents":["Deploy Dify as a containerized application in cloud environments","Configure different settings for development, staging, and production","Manage secrets and credentials securely across environments","Scale API and worker services independently"],"best_for":["Teams deploying Dify to Kubernetes or cloud container services","Organizations needing multi-environment configuration management","Developers running Dify locally with Docker Compose"],"limitations":["Docker images are large (>1GB); slow to build and push in CI/CD pipelines","Multi-stage builds require separate configuration for API and frontend; no unified build process","Environment variables are not validated at startup; misconfiguration may cause runtime failures","No built-in health checks or readiness probes; requires external orchestration (Kubernetes, Docker Swarm)"],"requires":["Docker 20.10+ for building and running containers","Docker Compose 2.0+ for local development stack","Kubernetes 1.20+ for production deployments (optional)","Environment variables for configuration (API keys, database URLs, etc.)"],"input_types":["Dockerfile specifications","Docker Compose YAML configuration","environment variable files (.env)","Kubernetes manifests (optional)"],"output_types":["Docker images (API, frontend)","running containers with exposed ports","deployment logs and status"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_11","uri":"capability://planning.reasoning.application.type.abstraction.supporting.chatbots.agents.and.workflows","name":"application type abstraction supporting chatbots, agents, and workflows","description":"Dify abstracts three Application Types (Chatbot, Agent, Workflow) with different execution models and capabilities. Chatbots use simple LLM calls with conversation history; Agents use ReAct-style reasoning with tool calling and multi-step planning; Workflows use explicit DAG execution with node composition. The Application Type determines available features (tool calling, knowledge retrieval, human input) and execution modes (streaming, async, batch).","intents":["Build simple chatbots without workflow complexity","Create autonomous agents with tool calling and reasoning","Design complex multi-step workflows with explicit control flow","Switch application types without re-implementing core logic"],"best_for":["Teams building multiple application types (chatbots, agents, workflows) in one platform","Developers prototyping different AI application architectures","Organizations standardizing on a single platform for AI application development"],"limitations":["Application type is immutable after creation; cannot convert chatbot to agent without re-creating","Feature availability varies by type; some features (e.g., human input) are workflow-only","Execution model differences (streaming vs async) require different client implementations","No unified testing framework across application types"],"requires":["Dify API server with application type support","PostgreSQL for application definition storage"],"input_types":["application type selection (Chatbot, Agent, Workflow)","application configuration (name, description, system prompt)","type-specific parameters (tools for agents, nodes for workflows)"],"output_types":["application ID and metadata","application execution results","type-specific outputs (conversation history, agent reasoning, workflow trace)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_12","uri":"capability://tool.use.integration.built.in.and.api.based.tool.integration.with.schema.validation","name":"built-in and api-based tool integration with schema validation","description":"Dify's Tool and Plugin Ecosystem supports three tool types: built-in tools (web search, calculator, etc.), API-based tools (HTTP requests with schema validation), and MCP tools (via MCP protocol). Tools are registered in a unified Tool Manager with JSON Schema definitions for parameter validation. When agents or workflows invoke tools, parameters are validated against schemas before execution, preventing invalid API calls and improving error handling.","intents":["Add web search, calculator, and other built-in tools to agents","Integrate custom APIs without writing tool adapters","Validate tool parameters before invocation to prevent errors","Support multiple tool types in a single agent or workflow"],"best_for":["Teams building agents with diverse tool requirements","Developers integrating APIs without custom adapter code","Organizations standardizing on schema-based tool integration"],"limitations":["Built-in tools are limited to web search and calculator; no extensibility for custom built-in tools","API-based tools require manual schema definition; no automatic schema generation from OpenAPI specs","Schema validation is strict; tools with optional parameters require careful schema design","No built-in retry logic for failed tool invocations; requires workflow-level error handling"],"requires":["Tool schemas defined in JSON Schema format","API endpoints for API-based tools","MCP servers for MCP tools (optional)"],"input_types":["tool definitions (name, description, schema)","tool invocation requests with parameters","tool configuration (API keys, endpoints)"],"output_types":["tool execution results","parameter validation errors","tool schema definitions"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_2","uri":"capability://memory.knowledge.knowledge.base.indexing.and.rag.pipeline.with.multiple.vector.database.backends","name":"knowledge base indexing and rag pipeline with multiple vector database backends","description":"Dify's Knowledge Base and RAG System manages document ingestion, chunking, embedding, and retrieval across multiple vector database backends (Pinecone, Weaviate, Qdrant, Milvus, etc.). The Document Indexing Pipeline processes uploaded files through a configurable chunking strategy, generates embeddings via provider-agnostic APIs, and stores vectors with metadata filtering. The RAG Pipeline Workflow retrieves relevant documents based on semantic similarity and metadata filters, then passes them to LLM nodes for context-aware generation.","intents":["Upload and index documents (PDF, DOCX, TXT, etc.) for semantic search","Retrieve relevant context from large document collections without token limits","Filter retrieval results by metadata (date, source, category) before passing to LLM","Switch vector database backends without re-indexing documents"],"best_for":["Teams building document-aware chatbots and Q&A systems","Organizations with large proprietary knowledge bases needing semantic search","Developers avoiding vendor lock-in by supporting multiple vector databases"],"limitations":["Chunking strategy is fixed per dataset; no dynamic chunk size adjustment based on query complexity","Metadata filtering is applied post-retrieval, not pre-filtering in vector DB queries, reducing efficiency for large datasets","No built-in support for hybrid search (keyword + semantic); requires custom retrieval nodes","Re-indexing after document updates requires full re-embedding, no incremental updates"],"requires":["At least one vector database configured (Pinecone, Weaviate, Qdrant, Milvus, etc.)","Embedding model API key (OpenAI, Hugging Face, etc.)","PostgreSQL for dataset metadata and indexing state","File storage backend (local filesystem, S3, etc.) for document persistence"],"input_types":["document files (PDF, DOCX, TXT, Markdown, CSV)","text queries for retrieval","metadata filters (key-value pairs)","chunking configuration (chunk size, overlap)"],"output_types":["indexed document chunks with embeddings","retrieved context passages","metadata-filtered results","indexing status and statistics"],"categories":["memory-knowledge","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_3","uri":"capability://tool.use.integration.mcp.protocol.integration.for.tool.and.plugin.execution","name":"mcp protocol integration for tool and plugin execution","description":"Dify integrates the Model Context Protocol (MCP) to enable dynamic tool and plugin discovery, schema registration, and execution. The MCP Client (SSE and streamable variants) communicates with MCP servers to fetch tool schemas, invoke tools with validated parameters, and handle streaming responses. Tools are registered in a unified Tool Manager that abstracts MCP, built-in, and API-based tools, allowing workflows to call external tools through a consistent interface without hardcoding tool implementations.","intents":["Dynamically discover and register tools from MCP servers without code changes","Execute external tools (APIs, scripts, services) from workflow nodes","Validate tool parameters against MCP schemas before invocation","Support streaming responses from long-running tool executions"],"best_for":["Teams building extensible AI agents with dynamic tool ecosystems","Developers integrating third-party services and APIs without custom adapters","Organizations standardizing on MCP for tool discovery and execution"],"limitations":["MCP server discovery is manual; no automatic service registry or health checking","Streaming responses from MCP tools require SSE support; not all MCP servers implement streaming","Tool parameter validation is schema-based but does not validate semantic correctness (e.g., valid API key format)","No built-in retry logic or circuit breaker for failed tool invocations"],"requires":["MCP server running and accessible (local or remote)","Python 3.9+ for MCP client implementation","Tool schemas defined in MCP server (JSON Schema format)"],"input_types":["MCP server connection details (URL, authentication)","tool invocation requests with parameters","tool schemas (JSON Schema)"],"output_types":["tool execution results","streaming response chunks","tool schema definitions","execution errors and status codes"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_4","uri":"capability://safety.moderation.multi.tenant.workspace.isolation.with.role.based.access.control","name":"multi-tenant workspace isolation with role-based access control","description":"Dify implements a Tenant Model with Resource Isolation that separates workspaces, datasets, workflows, and API keys by tenant. Role-Based Access Control (RBAC) enforces permissions at the workspace and member level, with roles (Admin, Editor, Viewer) controlling access to applications, datasets, and workflow execution. Authentication Methods support API keys, OAuth, and SAML, with Account Lifecycle Management handling user provisioning, deprovisioning, and workspace membership.","intents":["Create isolated workspaces for different teams or projects","Control who can edit workflows, access datasets, and deploy applications","Manage API keys per workspace with granular permission scopes","Audit user actions and enforce compliance policies"],"best_for":["SaaS platforms serving multiple organizations","Enterprise teams needing workspace isolation and audit trails","Developers building white-label AI platforms"],"limitations":["RBAC is workspace-level only; no fine-grained resource-level permissions (e.g., specific dataset access)","No built-in audit logging; requires external observability integration (Sentry, OpenTelemetry)","Workspace switching requires re-authentication; no seamless multi-workspace context","API key rotation is manual; no automatic key expiration or rotation policies"],"requires":["PostgreSQL for tenant and member data storage","OAuth provider (Google, GitHub, etc.) for SSO (optional)","SAML identity provider for enterprise SSO (optional)"],"input_types":["user credentials (email, password)","workspace creation requests","role assignments","API key generation requests"],"output_types":["authentication tokens (JWT)","API keys with scopes","workspace membership lists","audit logs"],"categories":["safety-moderation","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_5","uri":"capability://text.generation.language.chat.and.completion.api.with.conversation.history.and.feedback.tracking","name":"chat and completion api with conversation history and feedback tracking","description":"Dify exposes Chat and Completion APIs that accept user messages, route them through application logic (RAG, workflows, or simple LLM calls), and return responses with streaming support. Conversation history is persisted in PostgreSQL, enabling context-aware multi-turn interactions. Feedback APIs allow users to rate responses and provide corrections, which are stored for model fine-tuning and quality monitoring. The API Architecture uses request-response patterns with optional streaming via Server-Sent Events (SSE).","intents":["Build chatbot interfaces with multi-turn conversation support","Stream responses to users in real-time without waiting for full completion","Collect user feedback on AI responses for quality monitoring","Retrieve conversation history for audit and analysis"],"best_for":["Teams building chatbot UIs and conversational interfaces","Developers integrating Dify into existing applications via REST APIs","Organizations collecting feedback data for model improvement"],"limitations":["Conversation history is stored per application instance; no cross-application conversation sharing","Streaming responses require client-side SSE support; not compatible with simple HTTP polling","Feedback is stored but not automatically used for fine-tuning; requires external pipeline","No built-in rate limiting per user; requires external API gateway for production deployments"],"requires":["Dify API server running (Python 3.9+)","PostgreSQL for conversation and feedback storage","Client library or HTTP client for API calls"],"input_types":["user messages (text)","conversation context (previous messages)","application parameters (temperature, max_tokens, etc.)","feedback ratings and comments"],"output_types":["AI-generated responses (text)","streaming response chunks","conversation metadata (timestamps, tokens used)","feedback acknowledgment"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_6","uri":"capability://automation.workflow.visual.workflow.builder.with.drag.and.drop.node.composition","name":"visual workflow builder with drag-and-drop node composition","description":"Dify's Web Frontend Architecture provides a visual workflow builder UI built with Next.js that enables drag-and-drop composition of workflow nodes. The frontend renders Workflow Node UI Components for each node type (LLM, HTTP, Code, Knowledge Retrieval, Human Input), with real-time validation of node configurations and connections. The Chat Interface and Message Interactions component provides a conversational UI for testing applications and viewing multi-turn conversations with streaming message updates.","intents":["Design workflows visually without writing code","Test workflows interactively with real-time feedback","Configure node parameters through form-based UI","View workflow execution traces and debug failed runs"],"best_for":["Non-technical users building AI workflows","Teams prototyping workflows before production deployment","Developers iterating on workflow logic without code changes"],"limitations":["Visual builder does not support advanced node types (custom Python, complex branching logic)","Workflow validation is client-side only; server-side validation may catch errors not visible in UI","No collaborative editing; concurrent edits by multiple users may cause conflicts","Execution traces are limited to recent runs; historical traces require external logging"],"requires":["Node.js 18+ for frontend build","Modern web browser with JavaScript support","Dify API server for backend workflow execution"],"input_types":["node type selections","node configuration forms","workflow canvas interactions (drag, drop, connect)","test input messages"],"output_types":["workflow JSON definitions","rendered workflow diagrams","execution results with streaming updates","error messages and validation feedback"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_7","uri":"capability://data.processing.analysis.document.upload.and.file.management.with.format.conversion","name":"document upload and file management with format conversion","description":"Dify's File Upload and Management APIs handle document ingestion in multiple formats (PDF, DOCX, TXT, Markdown, CSV, JSON) with automatic format detection and conversion. Uploaded files are stored in configurable backends (local filesystem, S3, Azure Blob Storage) and associated with datasets for RAG indexing. The system extracts text and metadata from documents, applies document-level access controls, and tracks file lifecycle (upload, indexing, deletion).","intents":["Upload documents in various formats for knowledge base indexing","Extract text and metadata from PDFs and Office documents","Store files in cloud storage backends for scalability","Track document versions and manage file lifecycle"],"best_for":["Teams building document-aware AI applications","Organizations migrating legacy document repositories to AI platforms","Developers needing reliable file handling without custom storage logic"],"limitations":["Format conversion is lossy for complex documents (tables, images, formatting); plain text extraction only","File size limits are configurable but not enforced at API level; requires external gateway","No built-in virus scanning or malware detection; requires external security integration","Document versioning is not supported; re-uploading overwrites previous versions"],"requires":["File storage backend configured (local filesystem, S3, Azure Blob Storage, etc.)","Python 3.9+ for file processing","PostgreSQL for file metadata storage"],"input_types":["document files (PDF, DOCX, TXT, Markdown, CSV, JSON)","file metadata (name, description, tags)","storage backend configuration"],"output_types":["extracted text content","document metadata (pages, size, format)","file storage URLs","indexing status"],"categories":["data-processing-analysis","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_8","uri":"capability://automation.workflow.background.task.processing.with.celery.for.async.workflow.execution","name":"background task processing with celery for async workflow execution","description":"Dify uses Celery with Redis as a message broker to execute long-running workflows asynchronously in the background. Workflow execution requests are enqueued as Celery tasks, processed by worker processes, and results are stored in PostgreSQL for later retrieval. The Async Workflow Service manages task lifecycle (enqueue, execute, store results, handle failures), with support for task retries, timeouts, and dead-letter queues for failed tasks.","intents":["Execute long-running workflows without blocking API responses","Scale workflow execution across multiple worker processes","Retry failed workflow executions with exponential backoff","Monitor workflow execution status and retrieve results asynchronously"],"best_for":["Teams running workflows with long execution times (>30 seconds)","Organizations needing horizontal scaling of workflow execution","Developers building asynchronous AI pipelines"],"limitations":["Task state is not persisted across worker restarts; in-flight tasks may be lost","No built-in task prioritization; all tasks are processed FIFO","Celery monitoring requires external tools (Flower); no built-in task dashboard","Dead-letter queue handling is manual; failed tasks require manual inspection and retry"],"requires":["Redis server for message broker","Celery worker processes running (Python 3.9+)","PostgreSQL for task result storage","Dify API server for task enqueueing"],"input_types":["workflow execution requests","task parameters and context","retry configuration (max retries, backoff)"],"output_types":["task IDs for status polling","workflow execution results","task execution logs","failure reasons and stack traces"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-langgenius--dify__cap_9","uri":"capability://safety.moderation.observability.and.tracing.with.opentelemetry.and.sentry.integration","name":"observability and tracing with opentelemetry and sentry integration","description":"Dify integrates OpenTelemetry for distributed tracing and Sentry for error tracking and performance monitoring. The Trace Manager collects traces from workflow execution, LLM API calls, and database queries, exporting them to OpenTelemetry-compatible backends (Jaeger, Datadog, etc.). Sentry integration captures exceptions, performance metrics, and user feedback, enabling real-time alerting and post-mortem analysis of production issues.","intents":["Trace workflow execution across multiple services and identify bottlenecks","Monitor LLM API latency and token usage for cost optimization","Capture and alert on production errors in real-time","Analyze performance metrics and user feedback for quality improvement"],"best_for":["Teams running Dify in production with SLO requirements","Organizations needing detailed observability for cost and performance optimization","Developers debugging complex multi-service workflows"],"limitations":["Trace sampling is not configurable; all traces are exported, which may impact performance at scale","Sentry integration requires external Sentry account; no self-hosted error tracking","Trace data retention is limited by backend (Jaeger, Datadog); long-term analysis requires external storage","No built-in alerting rules; requires external monitoring platform (Datadog, New Relic, etc.)"],"requires":["OpenTelemetry-compatible backend (Jaeger, Datadog, etc.) for trace export","Sentry account for error tracking (optional)","Python 3.9+ for instrumentation"],"input_types":["workflow execution traces","LLM API call metrics","database query logs","exception stack traces"],"output_types":["distributed traces with timing","performance metrics (latency, throughput)","error reports with context","user feedback and annotations"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":44,"verified":false,"data_access_risk":"high","permissions":["API keys for at least one supported LLM provider (OpenAI, Anthropic, Gemini, etc.)","PostgreSQL or compatible database for quota state persistence","Python 3.9+ for backend API service","PostgreSQL for workflow definition and execution state storage","Redis for distributed task queue (Celery) if using background execution","Python 3.9+ for backend workflow engine","Docker 20.10+ for building and running containers","Docker Compose 2.0+ for local development stack","Kubernetes 1.20+ for production deployments (optional)","Environment variables for configuration (API keys, database URLs, etc.)"],"failure_modes":["Quota enforcement is soft-limit only — does not prevent requests already in-flight when quota exhausted","Provider-specific features (e.g., vision, function calling) require explicit adapter implementation per provider","No built-in fallback mechanism if primary provider quota is exhausted","DAG model does not support cycles or loops — workflows must be acyclic","Pause-resume state is stored in-memory by default; requires external persistence for multi-instance deployments","Node execution is sequential within a single workflow instance; no true parallel execution across nodes","Debugging paused workflows requires manual state inspection; no built-in debugger","Docker images are large (>1GB); slow to build and push in CI/CD pipelines","Multi-stage builds require separate configuration for API and frontend; no unified build process","Environment variables are not validated at startup; misconfiguration may cause runtime failures","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.4819765876840173,"quality":0.35,"ecosystem":0.7000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.3,"quality":0.25,"ecosystem":0.15,"match_graph":0.25,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.550Z","last_scraped_at":"2026-05-03T13:57:01.478Z","last_commit":"2026-05-03T13:51:47Z"},"community":{"stars":139956,"forks":21944,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=langgenius--dify","compare_url":"https://unfragile.ai/compare?artifact=langgenius--dify"}},"signature":"u5SJyeZbP0Ft+jdElbls/Ec2zinAQyeBcN4rZWP9FvQXJM3QOfF5GqZAbT7Q7wuYTG09rQ02BntRoJZbw5zjCQ==","signedAt":"2026-06-20T06:55:32.274Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/langgenius--dify","artifact":"https://unfragile.ai/langgenius--dify","verify":"https://unfragile.ai/api/v1/verify?slug=langgenius--dify","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"}}