{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github_mcp-awslabs-mcp","slug":"mcp-awslabs-mcp","name":"mcp","type":"mcp","url":"https://github.com/awslabs/mcp","page_url":"https://unfragile.ai/mcp-awslabs-mcp","categories":["mcp-servers"],"tags":["aws","mcp","mcp-client","mcp-clients","mcp-host","mcp-server","mcp-servers","mcp-tools","modelcontextprotocol"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github_mcp-awslabs-mcp__cap_0","uri":"capability://tool.use.integration.aws.service.tool.exposure.via.standardized.mcp.protocol","name":"aws service tool exposure via standardized mcp protocol","description":"Exposes 50+ AWS services (Lambda, DynamoDB, S3, CloudWatch, IAM, etc.) as callable tools through the Model Context Protocol, using a unified schema-based function registry that translates MCP tool definitions into AWS SDK calls. Each service gets a dedicated MCP server that implements the MCP specification's tools interface, allowing AI clients to discover and invoke AWS APIs with structured input/output validation without direct SDK knowledge.","intents":["I want my AI assistant to directly call AWS APIs without me writing custom integrations","I need to give Claude or other LLMs the ability to manage AWS resources through a standardized interface","I want to expose AWS operations as discoverable tools that any MCP-compatible client can use"],"best_for":["AI application developers building agents that need AWS resource access","DevOps teams automating infrastructure management through LLM interfaces","Enterprise teams standardizing AI tool access across multiple AWS services"],"limitations":["Requires AWS credentials (IAM role or access keys) configured in the MCP server environment","Tool discovery is limited to tools explicitly defined in each MCP server implementation — no dynamic AWS API introspection","Latency includes MCP protocol overhead plus AWS API call time, typically 500ms-2s per operation","No built-in request batching — each tool call results in a separate AWS API invocation"],"requires":["Python 3.9+","AWS credentials with appropriate IAM permissions for target services","MCP client implementation (Claude, Cursor, or custom MCP host)","Network access to AWS endpoints"],"input_types":["structured JSON parameters matching tool schema","AWS resource identifiers (ARNs, names, IDs)","configuration objects for service-specific operations"],"output_types":["JSON-structured responses from AWS APIs","error messages with AWS error codes","resource metadata and operation results"],"categories":["tool-use-integration","aws-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_1","uri":"capability://tool.use.integration.infrastructure.as.code.tool.integration.terraform.cdk.cloudformation","name":"infrastructure-as-code tool integration (terraform, cdk, cloudformation)","description":"Provides dedicated MCP servers for Terraform, AWS CDK, and CloudFormation that expose IaC operations as tools, enabling AI assistants to read, validate, plan, and apply infrastructure changes. The Terraform server parses HCL, the CDK server integrates with CDK CLI, and the CloudFormation server manages stack operations — each translating IaC-specific workflows into MCP tool schemas with structured input validation and change preview capabilities.","intents":["I want my AI assistant to generate and validate Terraform configurations before applying them","I need to let an LLM review infrastructure changes and explain what will be deployed","I want to automate CDK stack synthesis and CloudFormation deployments through an AI interface"],"best_for":["Infrastructure engineers using AI to accelerate IaC development","Platform teams building AI-driven infrastructure automation","Organizations standardizing infrastructure changes through LLM-assisted workflows"],"limitations":["Terraform server requires HCL parser and local Terraform binary — no remote state validation without explicit configuration","CDK server is TypeScript/Python-specific — cannot manage CDK projects in other languages","CloudFormation server cannot preview changes for nested stacks without explicit stack references","No built-in drift detection — requires separate CloudFormation drift detection API calls"],"requires":["Python 3.9+","Terraform binary (for Terraform server) or AWS CDK CLI (for CDK server)","AWS credentials with CloudFormation, IAM, and service-specific permissions","HCL parser library (terraform-parser or equivalent)"],"input_types":["HCL configuration files (Terraform)","CDK construct definitions (Python/TypeScript)","CloudFormation template JSON/YAML","variable overrides and parameter values"],"output_types":["terraform plan output (JSON format)","CDK synthesis output (CloudFormation template)","CloudFormation change set summaries","validation errors and warnings"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_10","uri":"capability://automation.workflow.mcp.server.lifecycle.management.and.process.orchestration","name":"mcp server lifecycle management and process orchestration","description":"Manages MCP server startup, shutdown, and communication through stdio, SSE (Server-Sent Events), or custom transports. The MCP host (client) spawns server processes, establishes bidirectional communication channels, handles connection lifecycle (initialization, heartbeats, graceful shutdown), and manages resource cleanup. This enables reliable server operation with automatic restart on failure and clean shutdown semantics.","intents":["I want to start and stop MCP servers as needed for my application","I need reliable communication with MCP servers including error handling and reconnection","I want to manage multiple MCP servers and route requests appropriately"],"best_for":["MCP host/client developers building server management infrastructure","Application developers deploying MCP servers in production","Teams building MCP orchestration platforms"],"limitations":["Process management is OS-specific — Windows and Unix have different process lifecycle semantics","No built-in load balancing — multiple server instances require external orchestration","Communication is synchronous — no built-in request queuing or async batching","Resource limits are not enforced — runaway servers can consume unbounded CPU/memory"],"requires":["Python 3.9+ (for Python MCP servers)","OS-level process management capabilities","Network stack for stdio/SSE communication"],"input_types":["server executable path and arguments","environment variables and configuration","transport type (stdio, SSE, custom)"],"output_types":["server process status (running, stopped, failed)","communication channel handles","error logs and diagnostics"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_11","uri":"capability://memory.knowledge.aws.documentation.and.developer.reference.access","name":"aws documentation and developer reference access","description":"Provides an MCP server that exposes AWS documentation and developer guides as searchable resources, enabling AI assistants to reference official AWS documentation without external web searches. The server indexes AWS docs and enables semantic search over documentation content, allowing AI to provide accurate, up-to-date information about AWS services, APIs, and best practices.","intents":["I want my AI assistant to reference official AWS documentation when explaining services","I need accurate API documentation without relying on external web search","I want to ensure AI recommendations align with AWS best practices"],"best_for":["AI application developers building AWS-focused assistants","Teams ensuring AI recommendations use official documentation","Organizations standardizing AWS knowledge through AI interfaces"],"limitations":["Documentation is static — updates require server restart or manual refresh","Search is limited to indexed content — no real-time documentation updates","No code example execution — documentation references only"],"requires":["Python 3.9+","AWS documentation index (pre-built or generated)","Search infrastructure (vector database or full-text search)"],"input_types":["search queries (natural language or keywords)","service names or API operations"],"output_types":["documentation excerpts with source URLs","API reference information","best practice recommendations"],"categories":["memory-knowledge","search-retrieval"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_2","uri":"capability://tool.use.integration.database.query.and.schema.introspection.postgresql.dynamodb.neptune.memcached","name":"database query and schema introspection (postgresql, dynamodb, neptune, memcached)","description":"Exposes database query execution and schema discovery as MCP tools through dedicated servers for PostgreSQL, DynamoDB, Neptune (graph), and Memcached. The PostgreSQL server uses SQLAlchemy for connection pooling and query execution with result streaming, DynamoDB server translates query patterns into DynamoDB API calls with scan/query optimization, and Neptune server handles Gremlin/SPARQL query execution — each providing structured schema introspection tools that allow AI assistants to understand data models before generating queries.","intents":["I want my AI assistant to query databases and understand schema structure without manual documentation","I need to let an LLM generate and execute SQL/DynamoDB queries with safety guardrails","I want to explore graph relationships in Neptune through natural language queries"],"best_for":["Data engineers building AI-assisted analytics and reporting","Application developers integrating database access into AI agents","Teams automating data exploration and query generation"],"limitations":["PostgreSQL server requires database connection pooling configuration — unbounded concurrent queries can exhaust connection limits","DynamoDB server cannot optimize complex queries across multiple tables without explicit index definitions","Neptune server requires Gremlin/SPARQL knowledge — natural language translation to graph queries is limited","No built-in query result pagination — large result sets must be streamed or limited by the MCP server","Schema introspection is read-only — no DDL operations (CREATE TABLE, ALTER) exposed as tools"],"requires":["Python 3.9+","Database credentials (connection string for PostgreSQL, IAM role for DynamoDB/Neptune)","SQLAlchemy (PostgreSQL), boto3 (DynamoDB/Neptune), or redis-py (Memcached)","Network access to database endpoints"],"input_types":["SQL queries (PostgreSQL)","DynamoDB query/scan parameters (table name, key conditions, filters)","Gremlin/SPARQL queries (Neptune)","Redis commands (Memcached)"],"output_types":["query result rows (JSON or CSV format)","schema metadata (table names, column types, indexes)","query execution statistics (rows scanned, execution time)","error messages with database-specific error codes"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_3","uri":"capability://tool.use.integration.container.and.kubernetes.orchestration.ecs.eks","name":"container and kubernetes orchestration (ecs, eks)","description":"Exposes container management operations through dedicated MCP servers for ECS (task definition management, service scaling, container logs) and EKS (pod management, deployment operations, cluster introspection). The ECS server translates tool calls into ECS API operations with task lifecycle management, while the EKS server uses kubectl or Kubernetes Python client to manage workloads, enabling AI assistants to deploy, scale, and troubleshoot containerized applications without direct CLI knowledge.","intents":["I want my AI assistant to deploy and scale ECS services based on metrics or user requests","I need to let an LLM manage Kubernetes deployments and troubleshoot pod issues","I want to automate container log retrieval and analysis through an AI interface"],"best_for":["DevOps engineers automating container deployments and scaling","Platform teams building AI-driven Kubernetes management","SRE teams using AI to troubleshoot container issues"],"limitations":["ECS server requires IAM permissions for ECS, EC2, and CloudWatch — no fine-grained per-service permissions","EKS server requires kubeconfig file or IAM role with EKS permissions — no in-cluster service account support","No built-in deployment rollback — requires explicit tool calls to revert to previous task definitions or deployments","Container log retrieval is limited to CloudWatch Logs (ECS) or kubectl logs (EKS) — no streaming or real-time tail support","Scaling operations are asynchronous — no built-in polling for completion status"],"requires":["Python 3.9+","AWS credentials with ECS/EKS permissions (for ECS/EKS servers)","kubectl binary and kubeconfig (for EKS server)","boto3 (ECS) or kubernetes Python client (EKS)"],"input_types":["ECS task definition names, service names, cluster names","Kubernetes deployment/pod names, namespaces","scaling parameters (desired count, replica count)","container image URIs"],"output_types":["task/pod status information (running, pending, failed)","container logs (CloudWatch Logs or kubectl logs output)","deployment/service metadata","scaling operation results"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_4","uri":"capability://tool.use.integration.ai.and.machine.learning.service.integration.bedrock.sagemaker.nova.canvas","name":"ai and machine learning service integration (bedrock, sagemaker, nova canvas)","description":"Exposes AWS AI/ML services as MCP tools through dedicated servers: Bedrock server provides access to foundation models and knowledge base retrieval, SageMaker server enables notebook execution and model training/inference, Nova Canvas server handles image generation and editing. Each server translates tool calls into service-specific APIs with streaming support for long-running operations, allowing AI assistants to invoke other AI models, retrieve knowledge, and generate content without direct SDK calls.","intents":["I want my AI assistant to invoke other foundation models (Claude, Llama, Mistral) through Bedrock","I need to let an LLM retrieve information from a Bedrock knowledge base without manual RAG implementation","I want to automate image generation and editing through Nova Canvas via an AI interface"],"best_for":["AI application developers building multi-model systems","Teams implementing AI-powered content generation pipelines","Organizations building AI agents that need access to specialized models"],"limitations":["Bedrock server requires model access permissions — not all models are available in all regions","Knowledge base retrieval is limited to Bedrock KB format — no integration with external vector databases","SageMaker server requires notebook instance or endpoint setup — no serverless notebook execution","Nova Canvas server is image-only — no video generation or 3D model support","No built-in cost tracking — requires separate Cost Explorer API calls to monitor spending"],"requires":["Python 3.9+","AWS credentials with Bedrock, SageMaker, or Nova Canvas permissions","boto3 with bedrock-runtime, sagemaker, and bedrock-agent-runtime clients","Bedrock model access enabled in AWS account"],"input_types":["text prompts (Bedrock, SageMaker)","image descriptions or editing instructions (Nova Canvas)","knowledge base query strings (Bedrock KB)","model parameters (temperature, max tokens, etc.)"],"output_types":["generated text (Bedrock models)","generated images (Nova Canvas)","knowledge base search results with citations","model inference outputs (SageMaker)"],"categories":["tool-use-integration","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_5","uri":"capability://tool.use.integration.monitoring.logging.and.observability.tool.access.cloudwatch.cloudtrail.cost.explorer","name":"monitoring, logging, and observability tool access (cloudwatch, cloudtrail, cost explorer)","description":"Exposes AWS monitoring and operational data as MCP tools through dedicated servers for CloudWatch (metrics, logs, alarms), CloudTrail (audit logs), and Cost Explorer (cost analysis). CloudWatch server provides metric queries and log insights execution, CloudTrail server enables audit log filtering and analysis, and Cost Explorer server translates cost queries into structured API calls — allowing AI assistants to analyze operational health, security events, and spending without manual dashboard navigation.","intents":["I want my AI assistant to analyze CloudWatch logs and metrics to diagnose application issues","I need to let an LLM query CloudTrail for security audits and compliance reporting","I want to understand AWS spending patterns through AI-assisted cost analysis"],"best_for":["SRE and DevOps teams automating incident investigation","Security teams building AI-assisted audit and compliance workflows","FinOps teams analyzing and optimizing cloud spending"],"limitations":["CloudWatch server requires CloudWatch Logs Insights syntax knowledge — natural language translation is limited","CloudTrail server is read-only — no event filtering or retention policy management","Cost Explorer server provides historical data only — no real-time cost tracking","Metric queries are limited to 1440 data points per request — long-term trend analysis requires multiple queries","No built-in alerting — requires separate CloudWatch Alarms API calls to create notifications"],"requires":["Python 3.9+","AWS credentials with CloudWatch, CloudTrail, and Cost Explorer permissions","boto3 with logs, cloudtrail, and ce (Cost Explorer) clients","CloudWatch Logs Insights query syntax knowledge (for advanced queries)"],"input_types":["log group names and query strings (CloudWatch Logs Insights)","metric names and dimensions (CloudWatch Metrics)","CloudTrail event filters (event names, resource types, time ranges)","cost analysis dimensions (service, linked account, region)"],"output_types":["log query results (JSON format)","metric data points with timestamps","CloudTrail event records with metadata","cost analysis summaries and trends"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_6","uri":"capability://tool.use.integration.identity.and.access.management.iam.policy.analysis.and.generation","name":"identity and access management (iam) policy analysis and generation","description":"Exposes IAM operations and policy analysis as MCP tools through a dedicated IAM server that can read existing policies, validate policy syntax, generate least-privilege policies, and simulate policy effects. The server uses IAM Policy Simulator API and policy document parsing to enable AI assistants to understand permission requirements, audit existing policies, and generate compliant IAM configurations without manual policy writing.","intents":["I want my AI assistant to analyze IAM policies and identify overly permissive access","I need to let an LLM generate least-privilege IAM policies for specific use cases","I want to simulate policy effects before applying them to production roles"],"best_for":["Security teams automating IAM policy review and compliance","DevOps engineers generating least-privilege policies for applications","Organizations standardizing IAM configurations through AI assistance"],"limitations":["Policy Simulator is limited to 1000 simulations per request — large-scale policy analysis requires batching","No support for resource-based policies (S3 bucket policies, KMS key policies) — only identity-based policies","Policy generation is template-based — cannot generate policies for custom or proprietary services","No built-in policy versioning — requires manual tracking of policy changes"],"requires":["Python 3.9+","AWS credentials with IAM read and policy simulator permissions","boto3 with iam and iam-policy-simulator clients"],"input_types":["IAM policy documents (JSON format)","principal ARNs (users, roles, service principals)","action names and resource ARNs for simulation","policy templates and requirements"],"output_types":["policy validation results (syntax errors, warnings)","policy simulator results (allowed, denied, implicit deny)","generated IAM policy documents","policy analysis summaries (permissions granted, potential risks)"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_7","uri":"capability://tool.use.integration.messaging.and.workflow.orchestration.sns.sqs.step.functions","name":"messaging and workflow orchestration (sns, sqs, step functions)","description":"Exposes message queue and workflow operations as MCP tools through dedicated servers for SNS (publish/subscribe), SQS (queue management), and Step Functions (workflow execution and monitoring). SNS server handles topic management and message publishing, SQS server manages queue operations and message retrieval, and Step Functions server enables workflow execution and state machine introspection — allowing AI assistants to orchestrate asynchronous workflows and integrate with event-driven architectures.","intents":["I want my AI assistant to publish messages to SNS topics and SQS queues","I need to let an LLM trigger and monitor Step Functions workflows","I want to automate event-driven workflows through an AI interface"],"best_for":["Application developers building event-driven AI workflows","DevOps teams automating asynchronous job orchestration","Teams integrating AI with message-driven architectures"],"limitations":["SQS server cannot guarantee message ordering — FIFO queue semantics require explicit configuration","Step Functions server requires state machine definition knowledge — no automatic workflow generation from natural language","Message publishing is fire-and-forget — no built-in delivery confirmation or retry logic","No built-in message filtering or routing — requires manual topic subscription management"],"requires":["Python 3.9+","AWS credentials with SNS, SQS, and Step Functions permissions","boto3 with sns, sqs, and stepfunctions clients"],"input_types":["SNS topic ARNs and message payloads","SQS queue URLs and message bodies","Step Functions state machine ARNs and execution inputs","message attributes and metadata"],"output_types":["message IDs and publish confirmations","queue message counts and attributes","Step Functions execution status and output","workflow execution history"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_8","uri":"capability://tool.use.integration.storage.and.data.access.s3.s3.tables.kendra.search","name":"storage and data access (s3, s3 tables, kendra search)","description":"Exposes object storage and search operations as MCP tools through dedicated servers for S3 (object operations, bucket management), S3 Tables (structured data queries), and Kendra (enterprise search). S3 server handles object upload/download/listing with metadata operations, S3 Tables server enables SQL-like queries on structured data, and Kendra server provides semantic search with document ranking — allowing AI assistants to access and search data without direct SDK knowledge.","intents":["I want my AI assistant to read, write, and list S3 objects","I need to let an LLM query structured data in S3 Tables using SQL","I want to search enterprise documents through Kendra without manual indexing"],"best_for":["Data engineers building AI-assisted data pipelines","Application developers integrating S3 storage into AI workflows","Enterprise teams implementing AI-powered document search"],"limitations":["S3 server cannot handle multipart uploads for large files — requires manual chunking for files >5GB","S3 Tables server requires table schema definition — no automatic schema inference","Kendra server is limited to indexed documents — no real-time indexing of new documents","No built-in access control — all operations use the MCP server's IAM role permissions","Object listing is paginated — large buckets require multiple queries"],"requires":["Python 3.9+","AWS credentials with S3, S3 Tables, and Kendra permissions","boto3 with s3, s3-tables, and kendra clients"],"input_types":["S3 bucket names, object keys, and file content","S3 Tables table names and SQL queries","Kendra index IDs and search queries","metadata and object attributes"],"output_types":["object content (binary or text)","object metadata (size, last modified, tags)","S3 Tables query results (JSON format)","Kendra search results with relevance scores and document excerpts"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-awslabs-mcp__cap_9","uri":"capability://tool.use.integration.mcp.server.discovery.and.capability.introspection","name":"mcp server discovery and capability introspection","description":"Provides mechanisms for MCP clients to discover available servers, introspect their exposed tools and resources, and understand tool schemas and capabilities. The MCP protocol defines initialization handshakes and tool listing endpoints that allow clients to query server metadata, tool definitions (with input/output schemas), and resource types — enabling dynamic client adaptation to available server capabilities without hardcoded tool knowledge.","intents":["I want my MCP client to automatically discover what tools are available from a server","I need to understand the input/output schemas for tools before calling them","I want to list all available resources (databases, files, etc.) that a server can access"],"best_for":["MCP client developers building dynamic tool discovery","AI application developers adapting to available server capabilities","Teams building MCP-compatible tools and frameworks"],"limitations":["Tool discovery is static — no runtime capability changes without server restart","Schema introspection is limited to JSON Schema format — no custom schema extensions","No capability versioning — breaking changes require client-side version negotiation","Resource listing can be expensive for servers with large resource counts — no pagination support"],"requires":["MCP client implementation","Network connectivity to MCP server","Understanding of MCP protocol initialization"],"input_types":["MCP server connection parameters (stdio, SSE, or custom transport)"],"output_types":["server metadata (name, version, capabilities)","tool definitions (name, description, input schema, output schema)","resource type definitions","capability flags (supports prompts, resources, etc.)"],"categories":["tool-use-integration","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":49,"verified":false,"data_access_risk":"high","permissions":["Python 3.9+","AWS credentials with appropriate IAM permissions for target services","MCP client implementation (Claude, Cursor, or custom MCP host)","Network access to AWS endpoints","Terraform binary (for Terraform server) or AWS CDK CLI (for CDK server)","AWS credentials with CloudFormation, IAM, and service-specific permissions","HCL parser library (terraform-parser or equivalent)","Python 3.9+ (for Python MCP servers)","OS-level process management capabilities","Network stack for stdio/SSE communication"],"failure_modes":["Requires AWS credentials (IAM role or access keys) configured in the MCP server environment","Tool discovery is limited to tools explicitly defined in each MCP server implementation — no dynamic AWS API introspection","Latency includes MCP protocol overhead plus AWS API call time, typically 500ms-2s per operation","No built-in request batching — each tool call results in a separate AWS API invocation","Terraform server requires HCL parser and local Terraform binary — no remote state validation without explicit configuration","CDK server is TypeScript/Python-specific — cannot manage CDK projects in other languages","CloudFormation server cannot preview changes for nested stacks without explicit stack references","No built-in drift detection — requires separate CloudFormation drift detection API calls","Process management is OS-specific — Windows and Unix have different process lifecycle semantics","No built-in load balancing — multiple server instances require external orchestration","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.6659729673471858,"quality":0.34,"ecosystem":0.6000000000000001,"match_graph":0.25,"freshness":0.75,"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":"2026-05-24T12:16:22.064Z","last_scraped_at":"2026-05-03T14:23:31.492Z","last_commit":"2026-05-03T01:40:08Z"},"community":{"stars":8938,"forks":1494,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=mcp-awslabs-mcp","compare_url":"https://unfragile.ai/compare?artifact=mcp-awslabs-mcp"}},"signature":"rzDKELfKW9A28IonylJsZmP+lgnuw5DdxC1iXipvrcAtaSix02eyaG15DtqCuU19CNbj4qNa187Xozu7OriNDQ==","signedAt":"2026-06-22T13:57:34.731Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/mcp-awslabs-mcp","artifact":"https://unfragile.ai/mcp-awslabs-mcp","verify":"https://unfragile.ai/api/v1/verify?slug=mcp-awslabs-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"}}