{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"npm_npm-restormelmcp","slug":"npm-restormelmcp","name":"@restormel/mcp","type":"mcp","url":"https://www.npmjs.com/package/@restormel/mcp","page_url":"https://unfragile.ai/npm-restormelmcp","categories":["mcp-servers"],"tags":["mcp","model-context-protocol","ai","restormel","tools"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"npm_npm-restormelmcp__cap_0","uri":"capability://tool.use.integration.mcp.tool.schema.generation.for.model.routing.and.selection","name":"mcp tool schema generation for model routing and selection","description":"Generates standardized MCP tool definitions that expose Restormel's model registry, enabling Claude and other MCP-compatible clients to query available models, their capabilities, pricing tiers, and routing rules. Uses JSON Schema to define tool inputs/outputs that map to Restormel's internal model metadata, allowing clients to make informed routing decisions without hardcoding model names.","intents":["Query available AI models and their capabilities from within an MCP client","Retrieve real-time pricing and cost information for different model providers","Determine which model to route a request to based on latency, cost, or capability requirements","Access model entitlements and usage limits for the current user or organization"],"best_for":["MCP client developers building model-agnostic AI applications","Teams using Claude with MCP to dynamically select between multiple LLM providers","Organizations managing multi-model inference with cost optimization requirements"],"limitations":["Requires Restormel backend to be running and accessible; no local fallback for model metadata","Schema-based tool definitions add discovery overhead on client startup","No built-in caching of model metadata — each query hits the Restormel service","Limited to models registered in Restormel's catalog; custom or private models require backend registration"],"requires":["Node.js 16+","MCP client implementation (Claude, or custom MCP-compatible tool)","Access to Restormel backend API endpoint","Valid authentication credentials for Restormel service"],"input_types":["JSON (MCP tool call parameters)","query strings (model name, provider filter, capability filter)"],"output_types":["JSON (model metadata objects)","structured pricing data","routing recommendations"],"categories":["tool-use-integration","model-routing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-restormelmcp__cap_1","uri":"capability://tool.use.integration.provider.abstraction.and.cost.calculation.via.mcp.tools","name":"provider abstraction and cost calculation via mcp tools","description":"Exposes Restormel's multi-provider abstraction layer as MCP tools, allowing clients to query cost estimates, token pricing, and provider-specific parameters (rate limits, context windows, supported features) without directly integrating with OpenAI, Anthropic, or other provider SDKs. Implements a unified cost calculation interface that normalizes pricing across providers with different billing models.","intents":["Calculate estimated cost for a request before routing to a specific model provider","Compare pricing across multiple providers for the same capability","Retrieve provider-specific constraints (context window, max tokens, rate limits)","Understand which providers support specific features (vision, function calling, streaming)"],"best_for":["Cost-conscious teams building multi-provider LLM applications","Developers building cost estimation or budget tracking features","Organizations with provider negotiation agreements needing custom pricing logic"],"limitations":["Cost calculations are estimates based on Restormel's pricing data; actual billing may vary with provider promotions or volume discounts","Requires Restormel to maintain up-to-date pricing for all supported providers; lag in price updates possible","No support for custom pricing models or internal cost allocation rules beyond what Restormel exposes","Provider-specific features (e.g., vision capabilities) depend on Restormel's metadata accuracy"],"requires":["Node.js 16+","MCP client with tool-calling support","Restormel backend with provider integrations configured","API credentials for at least one supported provider (OpenAI, Anthropic, etc.)"],"input_types":["JSON (model name, token count, feature flags)","query parameters (provider filter, cost threshold)"],"output_types":["JSON (cost estimates, pricing breakdown)","provider capability matrices","constraint specifications"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-restormelmcp__cap_2","uri":"capability://tool.use.integration.entitlements.and.usage.quota.enforcement.via.mcp.tools","name":"entitlements and usage quota enforcement via mcp tools","description":"Exposes Restormel's entitlements system as MCP tools, allowing clients to check user/organization quotas, usage limits, and feature access before making model requests. Implements a schema-based entitlements query interface that returns structured permission data (e.g., which models are available, rate limits, monthly token budgets) without requiring direct database access.","intents":["Check if a user has access to a specific model or provider before routing a request","Retrieve current usage metrics and remaining quota for the current billing period","Enforce rate limits or token budgets at the application level before hitting Restormel's backend","Determine which premium features are available based on the user's subscription tier"],"best_for":["Multi-tenant SaaS applications with per-user or per-organization quotas","Teams implementing client-side quota enforcement to reduce backend load","Organizations with tiered pricing models needing to gate features by subscription"],"limitations":["Entitlements are cached on the client; real-time quota updates require polling or webhook integration","No built-in support for complex entitlement rules (e.g., time-based access, conditional quotas); requires custom logic in Restormel backend","Quota enforcement is advisory — actual enforcement happens at Restormel backend; client-side checks can be bypassed","No support for cross-tenant quota sharing or pooled budgets"],"requires":["Node.js 16+","MCP client with tool-calling support","Restormel backend with entitlements system configured","User authentication and session management (API key or OAuth token)"],"input_types":["JSON (user ID, organization ID, feature name)","query parameters (quota type, billing period)"],"output_types":["JSON (entitlements object, quota limits, usage metrics)","boolean (access granted/denied)","structured permission matrix"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-restormelmcp__cap_3","uri":"capability://tool.use.integration.dynamic.documentation.and.capability.discovery.via.mcp.tools","name":"dynamic documentation and capability discovery via mcp tools","description":"Generates and exposes MCP tools that serve as dynamic documentation for Restormel's models, providers, and routing capabilities. Uses JSON Schema to define tool descriptions, parameters, and examples that Claude and other MCP clients can introspect to understand available models, their use cases, and best practices without requiring external documentation sites.","intents":["Discover available models and their recommended use cases directly within an MCP client","Retrieve detailed documentation for a specific model's capabilities and limitations","Access examples and best practices for using different models or providers","Get real-time updates to documentation without manual client updates"],"best_for":["Developers building AI applications who want in-context model documentation","Teams using Claude with MCP to explore Restormel's capabilities without external docs","Organizations maintaining multiple model versions and needing version-specific documentation"],"limitations":["Documentation is only as current as Restormel's backend; stale or incomplete metadata will propagate to clients","Schema-based documentation may be less detailed than hand-written docs; complex use cases may require external documentation","No support for rich formatting (markdown, images) in MCP tool descriptions; limited to plain text and JSON","Discovery requires active querying; no push-based notification when documentation changes"],"requires":["Node.js 16+","MCP client with introspection support (Claude 3.5+)","Restormel backend with documentation metadata configured","Network access to Restormel service"],"input_types":["query strings (model name, provider, use case filter)","JSON (documentation request parameters)"],"output_types":["JSON (model documentation objects)","markdown or plain text (capability descriptions, examples)","structured metadata (use cases, limitations, best practices)"],"categories":["tool-use-integration","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-restormelmcp__cap_4","uri":"capability://tool.use.integration.mcp.server.implementation.and.tool.registry.management","name":"mcp server implementation and tool registry management","description":"Provides a Node.js-based MCP server implementation that registers Restormel's tools (model routing, cost calculation, entitlements, documentation) and exposes them via the Model Context Protocol. Handles MCP protocol negotiation, tool schema validation, and request/response serialization, allowing any MCP-compatible client (Claude, custom agents) to call Restormel tools without implementing MCP protocol details.","intents":["Set up a Restormel MCP server that Claude or other clients can connect to","Register custom tools or extend Restormel's tool set with application-specific logic","Handle MCP protocol details (initialization, tool discovery, error handling) transparently","Deploy Restormel tools to a production environment with proper error handling and logging"],"best_for":["Teams deploying Restormel as a managed service for multiple MCP clients","Developers building custom MCP servers that need to integrate Restormel's capabilities","Organizations needing to extend Restormel's tools with custom business logic"],"limitations":["MCP server runs as a separate process; requires infrastructure to host and manage it","No built-in clustering or load balancing; single-instance deployments may become bottleneck","Tool registration is static at startup; adding new tools requires server restart","Limited observability; requires custom logging/monitoring integration for production use","No built-in rate limiting or request queuing; high-volume clients may overwhelm the server"],"requires":["Node.js 16+","npm or yarn package manager","@restormel/mcp package installed","Restormel backend API endpoint and credentials","MCP client library or compatible client (Claude, custom agent)"],"input_types":["MCP protocol messages (JSON-RPC)","tool call requests with parameters"],"output_types":["MCP protocol responses (JSON-RPC)","tool results (JSON, structured data)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":26,"verified":false,"data_access_risk":"high","permissions":["Node.js 16+","MCP client implementation (Claude, or custom MCP-compatible tool)","Access to Restormel backend API endpoint","Valid authentication credentials for Restormel service","MCP client with tool-calling support","Restormel backend with provider integrations configured","API credentials for at least one supported provider (OpenAI, Anthropic, etc.)","Restormel backend with entitlements system configured","User authentication and session management (API key or OAuth token)","MCP client with introspection support (Claude 3.5+)"],"failure_modes":["Requires Restormel backend to be running and accessible; no local fallback for model metadata","Schema-based tool definitions add discovery overhead on client startup","No built-in caching of model metadata — each query hits the Restormel service","Limited to models registered in Restormel's catalog; custom or private models require backend registration","Cost calculations are estimates based on Restormel's pricing data; actual billing may vary with provider promotions or volume discounts","Requires Restormel to maintain up-to-date pricing for all supported providers; lag in price updates possible","No support for custom pricing models or internal cost allocation rules beyond what Restormel exposes","Provider-specific features (e.g., vision capabilities) depend on Restormel's metadata accuracy","Entitlements are cached on the client; real-time quota updates require polling or webhook integration","No built-in support for complex entitlement rules (e.g., time-based access, conditional quotas); requires custom logic in Restormel backend","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.2,"ecosystem":0.45,"match_graph":0.25,"freshness":0.6,"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":"2026-05-24T12:16:24.482Z","last_scraped_at":"2026-05-03T14:23:57.392Z","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=npm-restormelmcp","compare_url":"https://unfragile.ai/compare?artifact=npm-restormelmcp"}},"signature":"cYH8Kx9iH7p/fljGP0xiYPj0YgrrNLQL5LhqzZ8fjxRsgCCoaxYc7sIXyzKL2GcQYF1pwkxWr4FlRdts3xoWAQ==","signedAt":"2026-06-21T10:16:06.838Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/npm-restormelmcp","artifact":"https://unfragile.ai/npm-restormelmcp","verify":"https://unfragile.ai/api/v1/verify?slug=npm-restormelmcp","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"}}