{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"npm-azure-mcp","slug":"azure-mcp","name":"@azure/mcp","type":"mcp","url":"https://github.com/Microsoft/mcp/blob/main/servers/Azure.Mcp.Server#readme","page_url":"https://unfragile.ai/azure-mcp","categories":["mcp-servers"],"tags":["azure","mcp","model-context-protocol"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"npm-azure-mcp__cap_0","uri":"capability://tool.use.integration.azure.resource.tool.exposure.via.mcp.protocol","name":"azure resource tool exposure via mcp protocol","description":"Exposes Azure cloud resources (compute, storage, networking, databases) as callable tools through the Model Context Protocol, enabling LLM agents to discover and invoke Azure operations via a standardized schema-based interface. Implements MCP's tool registry pattern to map Azure SDK operations into structured function definitions with JSON Schema validation, allowing Claude and other MCP-compatible clients to introspect available Azure capabilities and execute them with type-safe parameters.","intents":["I want my AI agent to directly provision and manage Azure infrastructure without writing custom integration code","I need to expose Azure operations as discoverable tools that Claude can call autonomously","I want to standardize how LLMs interact with Azure across multiple agents and applications"],"best_for":["DevOps teams building autonomous infrastructure management agents","Enterprise teams standardizing LLM-to-cloud integrations across Azure","Developers prototyping multi-step Azure workflows driven by LLM reasoning"],"limitations":["Requires Azure SDK authentication setup (connection strings, managed identities, or service principals) — no built-in credential management UI","Tool discovery is static at server startup — dynamic resource discovery requires custom extensions","Error handling delegates to Azure SDK exceptions — no unified error schema across different Azure services","No built-in rate limiting or quota management — relies on Azure service-level throttling"],"requires":["Node.js 18+ or Python 3.9+ (depending on SDK variant)","Azure subscription with appropriate RBAC permissions for resources being exposed","@azure/identity package for authentication","MCP-compatible client (Claude Desktop, custom MCP host, or compatible LLM framework)"],"input_types":["JSON-serialized Azure resource parameters","Azure resource IDs and names","Configuration objects matching Azure SDK schemas"],"output_types":["Azure resource objects (JSON)","Operation status and metadata","Resource IDs and connection strings","Structured error responses"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_1","uri":"capability://tool.use.integration.azure.authentication.and.credential.management.for.mcp.context","name":"azure authentication and credential management for mcp context","description":"Manages Azure authentication flows (service principals, managed identities, interactive login, connection strings) and injects credentials into the MCP server context so that tool calls execute with proper Azure authorization. Uses @azure/identity library's DefaultAzureCredential chain to support multiple authentication methods without code changes, automatically selecting the appropriate credential type based on the runtime environment (local development, container, managed identity).","intents":["I want my MCP server to authenticate to Azure without hardcoding credentials in the client","I need to support multiple authentication methods (dev, staging, production) with environment-based credential selection","I want to ensure all Azure tool calls are authorized with the correct identity and RBAC permissions"],"best_for":["Teams deploying MCP servers in containerized or serverless environments with managed identities","Multi-environment setups requiring credential rotation across dev/staging/prod","Organizations with strict credential management policies requiring non-interactive authentication"],"limitations":["DefaultAzureCredential chain has a fixed priority order — cannot customize credential selection strategy without forking","Interactive login (Device Code flow) requires user intervention, unsuitable for fully autonomous agents","No built-in credential caching or refresh token management — relies on @azure/identity's internal handling","Credential expiration is not surfaced to the MCP client — failed tool calls may occur silently if credentials expire mid-session"],"requires":["@azure/identity package (v3.0+)","Azure subscription with service principal or managed identity configured","Environment variables (AZURE_SUBSCRIPTION_ID, AZURE_TENANT_ID) or Azure CLI configuration for local development","Appropriate RBAC roles assigned to the authenticating identity"],"input_types":["Environment variables","Azure CLI configuration","Service principal credentials (optional, for explicit auth)"],"output_types":["Access tokens (internal, not exposed to client)","Authenticated Azure SDK clients ready for tool execution"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_10","uri":"capability://automation.workflow.azure.networking.and.security.configuration.management","name":"azure networking and security configuration management","description":"Exposes Azure Virtual Networks, Network Security Groups, Azure Firewall, and Application Gateway operations as MCP tools, enabling agents to configure network topology, security rules, and traffic management. Implements rule validation to prevent misconfiguration (e.g., overly permissive rules), supports network peering and VPN gateway setup, and provides network diagnostics tools for troubleshooting connectivity issues. Agents can define network policies declaratively and have the server translate them into Azure resource configurations.","intents":["I want my agent to configure network security rules and firewall policies without manual Azure Portal navigation","I need the agent to set up network peering and VPN connectivity between Azure resources","I want to diagnose and troubleshoot network connectivity issues automatically"],"best_for":["Network teams automating network configuration and security policy enforcement","Organizations implementing infrastructure-as-code for network topology","Teams building self-healing networks that detect and remediate connectivity issues"],"limitations":["Network rule validation is limited to syntax checking — semantic conflicts (e.g., overlapping CIDR ranges) may not be detected","Network changes have eventual consistency — rule changes may take 30-60 seconds to propagate","Network diagnostics (e.g., NSG flow logs) require separate configuration and storage account setup","Complex network topologies with many peering relationships can exceed LLM reasoning capabilities"],"requires":["@azure/arm-network SDK","Appropriate RBAC permissions for network resource management","Azure subscription with sufficient quota for network resources"],"input_types":["Network configuration parameters (CIDR ranges, subnets, routes)","Security rule definitions (source, destination, ports, action)","Peering and VPN gateway parameters"],"output_types":["Network resource IDs and metadata","Rule validation results","Network diagnostics (connectivity status, latency, packet loss)","Topology visualization data"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_2","uri":"capability://data.processing.analysis.azure.resource.enumeration.and.tool.schema.generation","name":"azure resource enumeration and tool schema generation","description":"Discovers available Azure resources and operations at server startup, dynamically generating MCP tool schemas that describe each Azure operation's parameters, return types, and documentation. Uses Azure SDK's type introspection and metadata to construct JSON Schema definitions for each tool, enabling MCP clients to understand what operations are available without hardcoding a tool catalog. Supports filtering and scoping to specific Azure services or resource groups to reduce tool surface area.","intents":["I want Claude to see all available Azure operations without me manually defining tool schemas","I need to limit which Azure resources and operations are exposed to the LLM based on security policies","I want tool documentation to automatically reflect Azure SDK changes without code updates"],"best_for":["Teams with large Azure estates wanting to expose broad infrastructure capabilities to agents","Security-conscious organizations needing fine-grained control over which operations are discoverable","Developers building multi-tenant MCP servers where different tenants need different tool subsets"],"limitations":["Schema generation is synchronous at startup — large Azure subscriptions with 1000+ resources may cause slow server initialization (30-60s)","Only exposes operations available in the Azure SDK — newer Azure features not yet in SDK may be unavailable","No built-in schema caching — every server restart regenerates all schemas, no incremental updates","Complex nested Azure resource types may generate deeply nested JSON schemas that exceed LLM context limits"],"requires":["Azure SDK for the services being enumerated (e.g., @azure/arm-compute, @azure/arm-storage)","Read permissions on Azure subscription to enumerate resources","Sufficient memory to hold schema definitions in memory (typically <50MB for standard subscriptions)"],"input_types":["Azure subscription ID","Resource group names (optional, for scoping)","Service filter list (optional, to limit which Azure services are exposed)"],"output_types":["MCP tool definitions (JSON Schema format)","Tool metadata (name, description, parameters, return type)"],"categories":["data-processing-analysis","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_3","uri":"capability://planning.reasoning.multi.step.azure.operation.orchestration.with.llm.reasoning","name":"multi-step azure operation orchestration with llm reasoning","description":"Enables LLM agents to compose multi-step Azure workflows by chaining tool calls across different Azure services, with the MCP server handling state management and dependency resolution between operations. The server maintains operation context across multiple tool invocations, allowing agents to reference outputs from previous steps (e.g., use a created VM's ID in a subsequent networking operation) without explicit state passing. Implements idempotency patterns to safely retry failed operations without duplicating resources.","intents":["I want my agent to provision a complete application stack (VMs, storage, networking, databases) in a single multi-step workflow","I need the agent to handle dependencies between Azure operations (e.g., create VNet before creating subnet)","I want failed operations to be safely retried without creating duplicate resources"],"best_for":["Teams automating complex infrastructure provisioning workflows","Developers building self-healing infrastructure agents that can recover from transient failures","Organizations standardizing infrastructure-as-code through LLM-driven orchestration"],"limitations":["State is ephemeral — stored in server memory only, lost on restart; requires external persistence for production use","No built-in transaction semantics — partial failures in multi-step workflows require manual rollback logic","Idempotency relies on Azure's eventual consistency model — race conditions possible in high-concurrency scenarios","Long-running operations (e.g., VM provisioning) may exceed LLM context windows if intermediate steps are verbose"],"requires":["MCP client capable of multi-turn conversations (Claude, custom agent framework)","Azure SDK clients for all services involved in the workflow","Appropriate RBAC permissions for all operations in the workflow chain"],"input_types":["Sequential tool calls from LLM agent","References to previous operation outputs (resource IDs, configuration objects)"],"output_types":["Workflow execution status","Intermediate operation results","Final provisioned resource metadata"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_4","uri":"capability://search.retrieval.azure.resource.monitoring.and.status.querying.via.mcp.tools","name":"azure resource monitoring and status querying via mcp tools","description":"Exposes Azure Monitor, Application Insights, and resource health APIs as MCP tools, enabling agents to query real-time metrics, logs, and status information about provisioned resources. Implements query builders that translate natural language monitoring requests into Azure Monitor KQL (Kusto Query Language) or REST API calls, returning structured time-series data and health status. Supports both synchronous status checks and asynchronous metric aggregation for long-running operations.","intents":["I want my agent to check if a newly provisioned VM is healthy before proceeding with configuration","I need the agent to query application logs and metrics to diagnose failures in deployed services","I want real-time visibility into resource utilization to inform scaling decisions"],"best_for":["Teams building self-healing infrastructure agents that monitor and respond to resource health","DevOps teams automating incident response workflows driven by monitoring data","Developers implementing cost optimization agents that analyze resource utilization"],"limitations":["Azure Monitor has a 1-2 minute ingestion latency — real-time monitoring is not possible, only near-real-time","KQL query complexity can exceed LLM reasoning capabilities — complex aggregations may require agent prompting or pre-built query templates","Metric retention varies by tier (7 days for basic, 93 days for standard) — historical analysis is limited","Application Insights requires explicit instrumentation — legacy applications without SDK integration cannot be monitored"],"requires":["Azure Monitor workspace or Application Insights instance configured","@azure/monitor-query SDK for KQL queries","Read permissions on monitoring data for the resources being queried"],"input_types":["Resource IDs or names","Metric names or log query strings","Time range specifications (e.g., 'last 1 hour')"],"output_types":["Time-series metric data (JSON)","Log entries (structured or raw text)","Resource health status (healthy, degraded, unavailable)","Aggregated statistics (min, max, avg, percentiles)"],"categories":["search-retrieval","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_5","uri":"capability://data.processing.analysis.azure.cost.analysis.and.resource.optimization.recommendations","name":"azure cost analysis and resource optimization recommendations","description":"Exposes Azure Cost Management APIs as MCP tools, enabling agents to analyze spending patterns, identify underutilized resources, and generate optimization recommendations. Implements cost aggregation across subscriptions and resource groups, supports filtering by service type or time period, and provides cost forecasting based on historical trends. Integrates with Azure Advisor to surface automated optimization recommendations (e.g., 'resize oversized VMs', 'delete unused storage accounts') as actionable tool outputs.","intents":["I want my agent to identify cost optimization opportunities and recommend resource changes","I need to analyze spending trends across multiple Azure subscriptions to understand cost drivers","I want the agent to automatically implement cost-saving changes (e.g., resize VMs, delete unused resources) based on recommendations"],"best_for":["FinOps teams automating cost optimization workflows","Multi-tenant organizations needing per-tenant cost visibility and chargeback","Developers building cost-aware infrastructure agents that balance performance with spending"],"limitations":["Cost data has a 24-48 hour lag — real-time cost tracking is not possible, only historical analysis","Recommendations from Azure Advisor are generic and may not account for application-specific requirements","Cost allocation across shared resources (e.g., shared storage accounts) requires manual tagging strategy","Forecasting accuracy degrades for new subscriptions with <30 days of historical data"],"requires":["Azure Cost Management + Billing enabled on subscription","@azure/arm-costmanagement SDK","Read permissions on billing and cost data","Azure Advisor enabled for optimization recommendations"],"input_types":["Subscription IDs or resource group names","Time range (e.g., 'last 30 days')","Filter criteria (service type, resource type, tag-based)"],"output_types":["Cost aggregations (total, by service, by resource)","Cost trends (daily/monthly breakdown)","Optimization recommendations (text descriptions + estimated savings)","Forecasted costs (projected spending for next month/quarter)"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_6","uri":"capability://safety.moderation.azure.key.vault.secret.and.certificate.management.via.mcp","name":"azure key vault secret and certificate management via mcp","description":"Exposes Azure Key Vault operations as MCP tools, enabling agents to securely manage secrets, certificates, and keys without exposing sensitive data to the LLM client. Implements secret versioning, rotation policies, and access control through Key Vault's RBAC model. Secrets are retrieved server-side and injected into Azure SDK clients or returned to the agent only when explicitly requested, ensuring sensitive data never flows through the LLM context.","intents":["I want my agent to retrieve database connection strings and API keys from Key Vault without exposing them in logs","I need to rotate secrets and certificates automatically based on policies","I want to audit all secret access and ensure compliance with security policies"],"best_for":["Security-conscious teams automating infrastructure provisioning with sensitive credential management","Organizations with strict compliance requirements (SOC 2, HIPAA) needing audit trails for secret access","DevOps teams implementing secret rotation and lifecycle management automation"],"limitations":["Key Vault has rate limits (2000 requests/10 seconds) — high-frequency secret access may be throttled","Secret retrieval requires network round-trip to Key Vault — adds latency to each tool call","No built-in secret masking in MCP responses — requires careful client-side handling to avoid logging sensitive data","Certificate renewal requires manual trigger or external automation — no built-in renewal workflow in MCP server"],"requires":["Azure Key Vault instance created and configured","@azure/keyvault-secrets and @azure/keyvault-certificates SDKs","RBAC permissions (Key Vault Secrets Officer, Key Vault Certificates Officer) on the vault"],"input_types":["Secret names or certificate names","Secret versions (optional, for retrieving specific versions)","Rotation policy parameters (for policy updates)"],"output_types":["Secret values (returned only when explicitly requested)","Certificate public keys and metadata","Rotation policy status","Access audit logs"],"categories":["safety-moderation","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_7","uri":"capability://safety.moderation.azure.role.based.access.control.rbac.policy.enforcement.and.auditing","name":"azure role-based access control (rbac) policy enforcement and auditing","description":"Enforces Azure RBAC policies at the MCP server level, ensuring that tool calls are authorized based on the authenticated identity's assigned roles. Implements policy evaluation by checking the identity's role assignments against required permissions for each tool, blocking unauthorized operations before they reach Azure. Logs all tool invocations with identity, operation, and result for audit compliance, supporting export to Azure Monitor or external SIEM systems.","intents":["I want to ensure the agent can only perform operations allowed by its Azure RBAC role","I need to audit all infrastructure changes made by the agent for compliance and troubleshooting","I want to prevent the agent from accessing resources outside its assigned scope (e.g., production resources)"],"best_for":["Enterprise teams with strict access control requirements and compliance mandates","Multi-tenant environments where agents must be scoped to specific subscriptions or resource groups","Organizations implementing least-privilege access models for infrastructure automation"],"limitations":["RBAC policy evaluation adds latency (~50-100ms per tool call) for permission checks","Azure RBAC has eventual consistency — newly assigned roles may not be reflected immediately in policy evaluation","Custom roles require manual definition — no automatic role generation from tool requirements","Audit logs are stored in Azure Monitor — requires separate configuration and retention policies"],"requires":["Azure RBAC roles assigned to the MCP server's identity","@azure/arm-authorization SDK for role assignment queries","Azure Monitor or external SIEM for audit log storage and analysis"],"input_types":["Tool name and parameters (for permission evaluation)","Resource IDs (for scope-based access control)"],"output_types":["Authorization decision (allowed/denied)","Audit log entries (identity, operation, timestamp, result)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_8","uri":"capability://automation.workflow.azure.deployment.and.infrastructure.as.code.template.execution","name":"azure deployment and infrastructure-as-code template execution","description":"Enables agents to deploy Azure resources using ARM templates or Bicep files through MCP tools, supporting parameterized deployments with validation and dry-run capabilities. Implements template validation before deployment to catch syntax errors and missing parameters, supports incremental and complete deployment modes, and tracks deployment status through Azure's deployment API. Agents can reference template outputs (e.g., deployed resource IDs) in subsequent tool calls.","intents":["I want my agent to deploy pre-defined infrastructure templates without manually constructing Azure API calls","I need to validate templates before deployment to prevent failed deployments","I want the agent to track deployment progress and handle failures gracefully"],"best_for":["Teams using ARM templates or Bicep as infrastructure-as-code and wanting LLM-driven deployments","Organizations standardizing infrastructure through template libraries","Developers building self-service infrastructure provisioning portals backed by agents"],"limitations":["Template validation is syntactic only — semantic errors (e.g., invalid parameter combinations) may only surface during deployment","Deployment rollback is not automatic — failed deployments require manual intervention or explicit rollback tool calls","Large templates (>4MB) may exceed Azure's deployment API limits","Template outputs are limited to 64KB — complex deployments with many outputs may be truncated"],"requires":["ARM templates or Bicep files accessible to the MCP server (local filesystem or URL)","@azure/arm-resources SDK for deployment operations","Appropriate RBAC permissions for all resources defined in templates"],"input_types":["Template file path or URL","Template parameters (JSON object)","Deployment mode (incremental or complete)","Resource group name"],"output_types":["Deployment validation results (success/failure with error details)","Deployment status (running, succeeded, failed)","Template outputs (resource IDs, connection strings, etc.)","Deployment history and change tracking"],"categories":["automation-workflow","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-azure-mcp__cap_9","uri":"capability://automation.workflow.azure.container.and.kubernetes.resource.management","name":"azure container and kubernetes resource management","description":"Exposes Azure Container Instances, Azure Container Registry, and Azure Kubernetes Service (AKS) operations as MCP tools, enabling agents to deploy, manage, and monitor containerized workloads. Implements container image management (push, pull, delete), AKS cluster operations (create, scale, upgrade), and pod/deployment management through Kubernetes API integration. Supports both imperative operations (create container, scale cluster) and declarative approaches (apply Kubernetes manifests).","intents":["I want my agent to deploy containerized applications to AKS without manual kubectl commands","I need the agent to scale Kubernetes deployments based on monitoring data or user requests","I want to manage container images in Azure Container Registry and deploy them to AKS automatically"],"best_for":["Teams running containerized workloads on AKS and wanting LLM-driven deployment automation","DevOps teams implementing self-healing Kubernetes clusters with autonomous scaling and remediation","Organizations standardizing container deployment through agents"],"limitations":["Kubernetes manifest complexity can exceed LLM reasoning capabilities — agents may struggle with complex YAML generation","AKS cluster creation is slow (10-15 minutes) — agents cannot wait synchronously for cluster readiness","Container image builds require external CI/CD pipeline — agents cannot build images directly","Kubernetes RBAC is separate from Azure RBAC — requires dual authentication (Azure credentials + Kubernetes service account)"],"requires":["Azure Container Registry instance or Docker Hub credentials","AKS cluster created and kubeconfig accessible to MCP server","@azure/arm-containerservice SDK for AKS operations","@kubernetes/client-node SDK for Kubernetes API access"],"input_types":["Container image names and tags","Kubernetes manifests (YAML or JSON)","Deployment parameters (replicas, resource limits, environment variables)","AKS cluster names and resource groups"],"output_types":["Deployment status (pending, running, failed)","Pod/container logs and events","Cluster scaling status","Image push/pull results"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":44,"verified":false,"data_access_risk":"high","permissions":["Node.js 18+ or Python 3.9+ (depending on SDK variant)","Azure subscription with appropriate RBAC permissions for resources being exposed","@azure/identity package for authentication","MCP-compatible client (Claude Desktop, custom MCP host, or compatible LLM framework)","@azure/identity package (v3.0+)","Azure subscription with service principal or managed identity configured","Environment variables (AZURE_SUBSCRIPTION_ID, AZURE_TENANT_ID) or Azure CLI configuration for local development","Appropriate RBAC roles assigned to the authenticating identity","@azure/arm-network SDK","Appropriate RBAC permissions for network resource management"],"failure_modes":["Requires Azure SDK authentication setup (connection strings, managed identities, or service principals) — no built-in credential management UI","Tool discovery is static at server startup — dynamic resource discovery requires custom extensions","Error handling delegates to Azure SDK exceptions — no unified error schema across different Azure services","No built-in rate limiting or quota management — relies on Azure service-level throttling","DefaultAzureCredential chain has a fixed priority order — cannot customize credential selection strategy without forking","Interactive login (Device Code flow) requires user intervention, unsuitable for fully autonomous agents","No built-in credential caching or refresh token management — relies on @azure/identity's internal handling","Credential expiration is not surfaced to the MCP client — failed tool calls may occur silently if credentials expire mid-session","Network rule validation is limited to syntax checking — semantic conflicts (e.g., overlapping CIDR ranges) may not be detected","Network changes have eventual consistency — rule changes may take 30-60 seconds to propagate","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.6489236966248564,"quality":0.32,"ecosystem":0.48999999999999994,"match_graph":0.25,"freshness":0.52,"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":"2026-05-24T12:16:23.328Z","last_scraped_at":"2026-05-03T14:04:47.472Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":87819,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=azure-mcp","compare_url":"https://unfragile.ai/compare?artifact=azure-mcp"}},"signature":"UI3tGm2xd/C6PbsvztWAy3W6Mb9SHYH1kXqfw0Q7CgjEweCUgQZTsEiilpwV9MiPkclCtoTSpN2i1MlorMX+Aw==","signedAt":"2026-06-21T13:31:47.123Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/azure-mcp","artifact":"https://unfragile.ai/azure-mcp","verify":"https://unfragile.ai/api/v1/verify?slug=azure-mcp","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"}}