mAIndala MCP Catalog

About

mAIndala MCP Catalog mAIndala is the open catalog of Model Context Protocol (MCP) services. This MCP server exposes 5 tools that let any AI agent instantly search, filter, and retrieve full metadata about 2,000+ MCP services — including endpoint URLs, authentication methods, transport protocols, tool schemas, configuration examples, and community ratings. --- ## Endpoint https://mcp.maindala.com/mcp Transport: StreamableHTTP (POST, responds as text/event-stream). No API key or session required. --- ## Configuring in Claude Desktop Add to ~/Library/Application Support/Claude/claudedesktopconfig.json: json { "mcpServers": { "maindala-catalog": { "command": "npx", "args": [ "mcp-remote", "https://mcp.maindala.com/mcp" ] } } } For MCP clients with native HTTP support: json { "mcpServers": { "maindala-catalog": { "url": "https://mcp.maindala.com/mcp" } } } --- ## Tools ### 1. getcataloginfo — Catalog Overview Returns a summary of the entire catalog: total service count, all categories with per-category counts, and supported protocol versions. Use this as a first call to orient yourself before searching. Input: (none) Example output: json { "totalServices": 1371, "categories": [ { "name": "AI & LLM Tools", "slug": "ai-llm-tools", "serviceCount": 1355 }, { "name": "Developer Tools", "slug": "developer-tools", "serviceCount": 3 } ], "description": "mAIndala is a catalog of MCP services that AI agents can connect to." } --- ### 2. listservices — Browse the Full Catalog Returns a paginated list of all services. Supports filtering by category and minimum rating. Use this when you want to discover services without a specific search term. Input parameters: | Parameter | Type | Default | Description | |---|---|---|---| | page | integer | 1 | Page number | | pagesize | integer | 20 (max 100) | Results per page | | category | string | — | Category name or ID | | minrating | number | — | Minimum aggregate rating (0–5) | Example: List the top-rated developer tools: json { "category": "developer-tools", "minrating": 4.0, "pagesize": 10 } --- ### 3. searchservices — Keyword & Semantic Search Full-text and vector-similarity search across service names, descriptions, and tags. The most powerful discovery tool — supports multiple simultaneous filters. Input parameters: | Parameter | Type | Description | |---|---|---| | query | string (required) | Search query | | category | string | Filter by category name or ID | | tags | string[] | Must match all listed tags | | protocolversion | string | e.g. "2025-03-26" | | authmethod | string | none, apikey, oauth2, jwt, basic | | minrating | number | Minimum aggregate rating | | pricingmodel | string | free, freemium, paid, enterprise | | page / pagesize | integer | Pagination | Example — find free GitHub tools with no auth required: json { "query": "github repository", "authmethod": "none", "pricingmodel": "free" } --- ### 4. getservice — Full Service Details Retrieves complete metadata for a single service by its UUID or slug. Returns the endpoint URL, auth method, transport, MCP config example, all registered tools with their input/output schemas, community ratings across 7 dimensions, and recent reviews. Input parameters: | Parameter | Type | Description | |---|---|---| | serviceid | string (required) | Service UUID or slug (e.g. "brave-search-mcp-server") | Returned fields include: - endpointUrl — The URL your agent should connect to - authMethod + mcpConfigExample — Ready-to-use configuration block - tools[] — Each tool's name, description, and full JSON input schema - aggregateRating + ratingBreakdown — Community ratings across: security, reliability, functional accuracy, documentation quality, ease of integration, performance/latency, and support/community - reviews[] — Text reviews from the community Example: Get full details for the Brave Search MCP Server: json { "serviceid": "brave-search-mcp-server" } --- ### 5. searchbycapability — Capability-Based Discovery Find services based on what they can do, using semantic similarity rather than exact keyword matching. Ideal when you know the capability you need but not the service name. Input parameters: | Parameter | Type | Description | |---|---|---| | capability | string (required) | Natural-language capability description | Example capabilities to search for: - "web search and browsing" - "image generation" - "read and write files" - "send email notifications" - "query SQL databases" - "interact with GitHub repositories" --- ## Typical Agent Workflows Workflow 1 — Discover a service for a specific task: 1. Call searchbycapability with a description of the capability needed 2. Call getservice on the best match to retrieve its endpoint URL and config example 3. Use the returned mcpConfigExample to connect to that service Workflow 2 — Find free, unauthenticated tools in a category: 1. Call getcataloginfo to see available categories 2. Call listservices with category + minrating filters 3. Call getservice on selected results to get full tool schemas Workflow 3 — Compare options for a known task: 1. Call searchservices with a precise query and authmethod: "none" filter 2. Review aggregateRating and ratingBreakdown fields across results 3. Select the service with the highest combined ease-of-integration and reliability scores --- ## Protocol - MCP Protocol Version: 2024-11-05 - Transport: StreamableHTTP (POST /mcp, responses as text/event-stream) - Authentication: None required - Pricing: Free, no rate limits - Source: Open catalog — community-submitted and auto-indexed from public registries

Available Tools(5)