# Vinkius AI Catalog MCP for AI Agents AI Agent Connect > Vinkius AI Catalog lets your AI agent explore the largest curated collection of Connectors directly. It can search for specific integrations, browse by category, check debugger grades, and read documentation or prompt examples programmatically. It's the way your agent finds the tools it needs to actually do work. ## Overview - **Category:** growth-engine - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_mgLuRhIN64QtERLUltG0fhiWwQraI7y7TZ9s8l3O/ai-agent-connect - **Tags:** vinkius, mcp-catalog, integration-discovery, ai-agents, debugger ## Description You want your agent to be able to find the right tools without you having to copy-paste lists of links or documentation into a prompt. This Connector lets your agent browse the entire Vinkius marketplace on its own. Instead of you doing the research, your agent can look up specific categories, check which Connectors have the best debugger grades, and read through FAQs or prompt examples to understand how to use an integration before it ever tries to run a command. It turns the catalog into a searchable database for your agent. You can have it monitor how many new integrations are being added or find specific tags to see what's trending. It's about giving your agent the ability to self-serve its own toolset. Because Vinkius hosts thousands of integrations, this connection ensures your agent always knows what's available and how to use it properly. ## Tools ### get_credentials_schema Returns credential_schema — the field keys, their types and which are required — plus credentials_configured (whether they are already set). Construct the set_credentials payload from these keys. Secret values are never returned. Get the credential fields an MCP requires. Call this to learn the exact schema you must fill before calling set_credentials ### get_listing_prompts Get prompt examples for a specific MCP to see how to interact with it. This provides your agent with real-world context. ### get_listing Get the full details of an Connector including its tools, FAQs, and related listings. This provides a complete profile for your agent. ### get_manifest Returns each MCP with a tools[] array of {name, description, inputSchema} (JSON Schema). Fetch the callable tools and their input schemas for one or more of the user MCPs. Call before invoking a downstream MCP tool so you know its exact arguments ### search_catalog Search the catalog using natural language or specific keywords. Your agent can use this to find tools based on a specific task. ### set_credentials Handles both owned and installed MCPs. Save the credentials an MCP needs so its tools become usable. Call get_credentials_schema first to learn the required fields ### active_mcps Narrow with filter (active|inactive), search (by name) or page. Enumerate the MCPs the user already has. Call this first to obtain the MCP ids needed by get_manifest, credentials_status and set_credentials ### credentials_status Returns credentials_configured (boolean) and oauth_status when applicable. For the field keys to fill, use get_credentials_schema instead. Check whether an MCP credentials are already set (and, for OAuth MCPs, whether it is connected). Use to decide if set_credentials is still needed ### get_listing_faqs Get the FAQs for a specific MCP to understand its common use cases. This helps your agent troubleshoot common issues. ### get_random_listings Get a rotating set of random Connectors from the catalog for discovery. It's a good way to find tools you didn't know existed. ## Prompt Examples **Prompt:** ``` Find me some high-rated tools for social media management. ``` **Response:** ``` | Category | MCP Name | Debugger Grade | Tools | | --- | --- | --- | --- | | Social Media | Buffer Integration | A+ | 12 | | Social Media | LinkedIn Connector | A | 8 | | Social Media | Twitter Pro | B | 5 | I found these high-rated options for you. Would you like to see the specific tools for any of these? ``` **Prompt:** ``` What are some examples of how to use the Kaggle Market Intelligence MCP? ``` **Response:** ``` Here are some prompt examples for that Connector: * **Prompt:** "Search for the latest datasets on competition trends." * **Response:** "I've scanned the Kaggle database and found 15 new datasets related to market trends." This shows how the agent handles queries for that specific integration. ``` **Prompt:** ``` Give me a summary of what's currently in the Growth Engine category. ``` **Response:** ``` The Growth Engine category currently features 22 active Connectors. This category includes: * **SEO Tools** * **Email Marketing Connectors** * **Lead Gen Automations** Would you like me to list the top 5 most popular ones in this category? ``` ## Capabilities ### Search the catalog with natural language Your agent can find specific Connectors by asking questions in plain English. ### Browse filtered categories Navigate the marketplace by specific tags, categories, or integration types. ### Inspect full MCP profiles Get deep details on tools, debugger grades, and trust scores for any listing. ### Retrieve prompt examples Pull real-world prompt and response pairs to see how to interact with a tool. ### Monitor catalog growth Check real-time stats to see how many Connectors are currently available. ### Get integration FAQs Access common questions and answers for specific Connectors to help with troubleshooting. ## Use Cases ### Finding the right CRM tool You don't know which CRM tool to use. Ask your agent to search the catalog for CRM integrations and compare their tool counts. ### Learning interaction styles You need to build a workflow but don't know the prompt style. Use get_listing_prompts to see how other users are interacting with a specific MCP. ### Daily tool discovery You want to see what's new. Have your agent run get_random_listings to suggest new tools to try every day. ### Auditing the tool stack You need to audit your current stack. Use get_all_categories to see every available category and find gaps in your current toolset. ## Benefits - Your agent can find specific tools using natural language with search_catalog. - You can audit the entire ecosystem for specific tags using get_all_tags. - Agents can learn how to use tools by reading prompt examples with get_listing_prompts. - You can see which tools are most reliable by checking debugger grades in get_listing. - Your agent can stay updated on new tools by checking get_catalog_stats. - You can quickly identify the best integrations by using get_featured_categories. ## How It Works The bottom line is your agent gains the ability to autonomously discover and evaluate the best tools for any task. 1. Connect this Connector to your Vinkius dashboard. 2. Your agent accesses the public catalog without needing extra credentials. 3. Ask your agent to find a specific tool or category, and it returns the full profile. ## Frequently Asked Questions **Can the Vinkius AI Catalog help my agent find specific tools?** Yes, it lets your agent search the entire catalog using natural language or specific keywords to find exactly what it needs. **How does the Vinkius AI Catalog handle tool reliability?** It provides debugger grades and trust scores for every Connector, so your agent can prioritize the most reliable integrations. **Can my agent read the documentation for an Connector it finds?** Yes, it can pull the full Markdown documentation and FAQs for any listing in the catalog. **Is there a way for my agent to see what's new in the Vinkius catalog?** Your agent can check the latest statistics and browse the most recent listings to stay updated on new tools. **Can I use the Vinkius AI Catalog to see how to write prompts for other Connectors?** Yes, the Connector can retrieve specific prompt examples for any integration to help your agent learn the best way to interact with it. **Does the Vinkius AI Catalog support browsing by category?** It lets your agent navigate the full taxonomy of featured and organic categories with real-time listing counts. **Do I need credentials to use this Connector?** No. The Vinkius catalog is fully public. All search, browse, and detail operations work without any credentials. **What data does the search return?** Search returns MCP cards with title, description, slug, tool count, debugger quality grade, and auth requirement. Results are ranked by relevance using AI-powered keyword extraction. **Can I filter by category and server type simultaneously?** Yes. All filter parameters (category, tag, server_type) can be combined in both search and browse operations for precise results. **What does the debugger grade mean?** Each Connector is scanned by the Vinkius Debugger, which grades quality from A (excellent) to F (critical issues). The grade reflects tool schema compliance, error handling, response consistency, and documentation completeness. **Do I need credentials to use this MCP?** No. The Vinkius catalog is fully public. All search, browse, and detail operations work without any credentials.