> ## Documentation Index > Fetch the complete documentation index at: https://docs.tydli.io/llms.txt > Use this file to discover all available pages before exploring further. # Real-World Use Cases > Practical applications of Tydli MCP servers # Real-World Use Cases Discover how teams are using Tydli to connect AI agents with their APIs and workflows. ## Customer Support AI **Connect Claude to your CRM, ticketing system, and knowledge base.** ### The Scenario Your support team handles hundreds of tickets daily. You want AI to help agents find customer information, create tickets, and suggest solutions faster. ### How It Works Deploy your CRM API, ticketing system API, and knowledge base API as MCP servers with Tydli. Connect them all to Claude Desktop. ### Example Interaction > **Agent**: "Can you look up customer [john@example.com](mailto:john@example.com) and show me their recent tickets?" > > **Claude**: *Uses your MCP servers to:* > > 1. Search CRM for customer [john@example.com](mailto:john@example.com) > 2. Retrieve customer ID and account details > 3. Query ticketing system for recent tickets > 4. Format and present the information Claude can search customer records, create support tickets, and suggest solutions from your documentation - all through natural conversation. ### APIs You Might Use * Zendesk, Intercom, or custom ticketing system * Salesforce, HubSpot, or custom CRM * Confluence, Notion, or custom knowledge base *** ## Development Workflow **Integrate GitHub, Jira, and CI/CD pipelines with AI agents.** ### The Scenario Developers spend significant time context-switching between GitHub, Jira, and CI/CD dashboards. You want AI to help manage the entire workflow from a single interface. ### How It Works Deploy GitHub API, Jira API, and your CI/CD platform API as MCP servers. Connect to Claude Desktop or Continue.dev in VS Code. ### Example Interaction > **Developer**: "Create a new branch, list open PRs, and check build status" > > **Claude**: *Uses your MCP servers to:* > > 1. Create a new Git branch via GitHub API > 2. List all open pull requests > 3. Check CI/CD pipeline status > 4. Report back with comprehensive status It handles all three APIs seamlessly with context awareness. ### APIs You Might Use * GitHub, GitLab, or Bitbucket * Jira, Linear, or Asana * CircleCI, Jenkins, or GitHub Actions *** ## Business Intelligence **Let AI query your analytics, databases, and reporting tools.** ### The Scenario Your team needs to pull reports from multiple data sources. Creating custom SQL queries or navigating complex BI tools takes too much time. ### How It Works Deploy your analytics platform API, database API, and reporting tool API as MCP servers. Give ChatGPT or Claude access. ### Example Interaction > **Analyst**: "What were our top 5 products by revenue last quarter? Compare to previous quarter." > > **ChatGPT**: *Uses your MCP servers to:* > > 1. Query sales database for Q3 2024 revenue > 2. Query sales database for Q2 2024 revenue > 3. Calculate product rankings and comparisons > 4. Format results with insights ChatGPT can pull sales data, compare metrics across periods, and generate insights without writing SQL. ### APIs You Might Use * Snowflake, BigQuery, or PostgreSQL * Tableau, Looker, or Power BI * Google Analytics, Mixpanel, or Amplitude *** ## E-commerce Assistant **Connect Shopify, inventory systems, and payment processors.** ### The Scenario You run an e-commerce business and want to provide customers with an AI-powered shopping assistant that can check inventory, process orders, and handle returns. ### How It Works Deploy your e-commerce platform API, inventory management API, and payment processor API as MCP servers. Build a custom AI assistant powered by Claude or ChatGPT. ### Example Interaction > **Customer**: "Do you have the blue widget in size large? If so, I'd like to order 2." > > **AI Assistant**: *Uses your MCP servers to:* > > 1. Check inventory for "blue widget size large" > 2. Verify stock availability (2+ in stock) > 3. Create order in e-commerce system > 4. Process payment via payment API > 5. Send confirmation with tracking info An AI-powered shopping assistant that checks inventory, processes orders, and handles returns through your APIs - completely automated. ### APIs You Might Use * Shopify, WooCommerce, or custom e-commerce * ShipStation, EasyPost, or custom fulfillment * Stripe, PayPal, or Square *** ## Additional Use Cases ### Marketing Automation Connect your marketing stack (email, CRM, analytics) to create campaigns, segment audiences, and analyze performance through AI. ### Financial Operations Integrate accounting software, payment processors, and banking APIs for automated reconciliation and reporting. ### HR & People Operations Connect HRIS, payroll, and benefits systems to answer employee questions and automate onboarding. ### Content Management Link CMS, DAM, and publishing tools to manage content creation, editing, and distribution workflows. ### DevOps & Monitoring Integrate monitoring tools, cloud platforms, and incident management for AI-assisted operations. ## Building Your Own Use Case ### 1. Identify Your APIs What systems do you use regularly that have APIs? * SaaS platforms (Salesforce, Slack, etc.) * Internal services (databases, microservices) * Third-party integrations (payment processors, shipping) ### 2. Check for OpenAPI Specs Most modern APIs provide OpenAPI/Swagger specifications: * Check API documentation for "OpenAPI", "Swagger", or "API Spec" * Look for `.json` or `.yaml` spec files * Use tools to generate specs from API docs if needed ### 3. Deploy with Tydli 1. Create a new deployment in Tydli 2. Upload your OpenAPI spec 3. Configure authentication 4. Get your MCP server URL ### 4. Connect to Your AI * Claude Desktop: Add to `claude_desktop_config.json` * ChatGPT: Use custom GPT with MCP support * Custom app: Use MCP client libraries ### 5. Test and Iterate * Start with simple queries * Expand to complex workflows * Monitor usage and performance * Refine your OpenAPI spec for better AI understanding ## Need Inspiration? Check out our community showcase: * Browse the [Tydli Explore page](https://tydli.io/explore) for public MCP servers * Join our [Discord](https://discord.gg/tamXgrqxAF) to see what others are building * Share your use case and get feedback from the community ## Next Steps * Start with our [Quickstart Guide](/getting-started/quickstart) to deploy your first MCP server * Review [Best Practices](/guides/best-practices) for production deployments * Learn about [Authentication Methods](/guides/authentication-methods) for secure integrations