Thanks to a great idea from Ray Deck, I have reduced the number of tools in the snappy.ai Xano MCP from ~180 tools and ~150,000 tokens to ~1000 tokens and 3 core tools: 1. Gemini Flash Powered Search - "What tools do I have?" (Semantic search across all tools - e.g., "create table", "list functions") 2. Info - "Tell me more about that specific tool" (returns parameters, examples, docs) 3. Execute - "Do it" (runs the tool with the right parameters) Now the initial handshake costs only ~1000 tokens instead of 150,000. Claude discovers tools as needed rather than loading everything upfront. This pattern should work for any MCP with dozens or hundreds of tools. Instead of exposing everything at once, create a search/discovery layer that lets Claude Code find what it needs when it needs it!

Hey, I'm working in n8n and trying to pass an array of values to the JSON body of the HTTP Request node, but I keep getting a comma-separated string instead of an actual array. I managed to fix it by adding square brackets and using .map() to wrap each item in quotes: "keywords": [{{ $json.output.keywords.map(k => `"${k}"`) }}] It works, but I'm wondering — is there a simpler or more elegant way to achieve this?

I spent the last few days at a cabin, taking time to put together some thoughts. When I started this community in November 2022, 'communities' weren't a popular thing yet. It was just me, making posts nobody read, hosting weekly meetings that no one showed up to. I was miles away from product-market fit, but I was carefully listening to the signs this group was pointing me towards. @Rene Tatua was the first person to join the weekly call consistently, and when you're dancing alone, you need to have at least one person to join you so that you stop looking crazy. Eventually, @Mario Haarmann joined too, and the three of us have kept 'Happy Friday' going ever since. I used to stress about what to do with this community. As trends came and went, I admit I explored some of them - making the group paid, running affiliates, or starting another 'premium' group. I've had plenty of people offer to partner with me in order to monetize it, but we had already built something valuable by simply meeting every Friday. I'm thankful for the genuine group we have. While my business continues to grow and change, our calls have kept me grounded, and served as a reminder me that huge value can come from ignoring trends, and staying consistent.

I'm excited to announce my Xano MCP is now live on Smithery and ready for everyone to use! After a week of refinement, it offers a comprehensive suite of 43 tools to interact with your Xano instances. What Can This MCP Do? With this toolset, you can: - Manage Xano instances with ease - Create, update, and search databases and tables - Build complete data schemas and handle records - Perform essential file operations - Design API groups and individual endpoints - Export workspaces for backup or transfer The MCP streamlines Xano development, helping you build faster and more efficiently while maintaining full control over your instances. Check out the link in the comments to install it directly from Smithery. Whether you're a Xano veteran or just getting started, these tools will significantly enhance your development workflow. Let me know if you have any questions or feature requests!

In the traditional agency model, you eventually reach a phase of the engagement where the core app has been built, and further features will be billed at an hourly rate for a dev managed by the agency. That hourly rate can be double or triple what the dev is receiving - because the agency will provide management services or some other kind of support alongside those hours. But by the agency charging rent on every hour that dev works, they become a pebble in the client's shoe, making it that much harder for them to accomplish their goal of building a successful product. It is unlikely that they are providing value on every hour that the dev works, to where a 2-3X multiple is justified. If our north star is the client's success, then it seems that a more favorable arrangement can be made; the client working directly with the dev, without the agency's involvement. For the client, they get a dev who is already trained, and the cost of hiring them just went down. For the dev, they get a client they are familiar with, and a raise relative to what the agency was paying them. For the agency, the amount of hours doing management work just went down, and the amount of hours doing high impact work just went up. You have allowed yourself to step out of where you are not adding much value, and step into areas where you are most needed. Areas such as training the client and their team in how to manage Stripe subscriptions themselves. Showing their UI/UX designer how to go straight from Figma to WeWeb, streamlining their development process. Workshops for their backend developer in how to use Xano, so they can bind data directly to the front end and eliminate bottlenecks. You can also continue to provide management to the dev, according to a set schedule. The agency becomes a high-impact catalyst, enabling the client to stand on their own two feet faster and more effectively. By focusing on empowerment rather than dependence, you can create a win-win-win scenario for the client, the dev, and the agency. The client gets a successful product, the dev gets on-going work, and the agency gets to make a real difference while maintaining healthy margins.