AI agents can call APIs, but can they actually understand them?
@Pascal_Senn shows why #GraphQL is such a strong fit for agents when using semantic introspection for huge production schemas.
This one is worth watching 👇
https://t.co/DhSb9JsDnW
With Fusion & HotChocolate 16, we're launching a new MCP adapter to expose GraphQL operations as MCP tools. We're also adding support for the MCP Apps standard for building agentic UI components. Check out the latest post for all the details: https://t.co/K4jxZAIUEY
Hot Chocolate 16 is here 🚀
- Implemented the GraphQL error mode proposal
- Added support for the new incremental delivery response format
- Implemented the new GraphQL batching proposal
- All core scalars are now specified on https://t.co/9yWBk9HUYW
Read more 👇
https://t.co/THOM84HHWS
#dotnet #graphql #aspnetcore
That’s a wrap on GraphQL Conf 🎉
2 days, 3 talks and 1 keynote from Michael Staib and Pascal Senn.
Then 1 day Unconf working group at Meta.
Thanks to everyone who joined, asked sharp questions, and pushed the conversation
#GraphQLConf#GraphQL
Day 2 at #GraphQLConf and Pascal just dropped some 🔥
Production GraphQL schemas are too big for context windows and agents struggle with them. Semantic introspection could change that: agents navigate schemas and interact with any GraphQL API reliably.
#GraphQL#AI#Agents
@nickchapsas You can build commercial success around OSS without slapping a license onto your OSS project. Thats why #HotChocolate will always stay MIT.
#HotChocolate#graphql will ALWAYS stay MIT. We have built with @Chilli_Cream a company around OSS that now employs 8 full-time engineers and we are growing. We are 100% self funded and have no pressure to squeeze our users.
https://t.co/cRKPP0tc8Y
@davidfowl I launched @blazorsaas recently exactly for that. It ticks all those boxes, pretty much. MartenDB for persistence, @Chilli_Cream GraphQL api and client, and Blazor frontend + Stripe, PostMark, nuke build, docker support and deployment to Coolify
https://t.co/WNfNIMyPjb
We have a new @graphql foundation youtube channel where we will publish #graphql conf and user group content. Make sure to subscribe to keep up to date with whats new in GraphQL. 🚀
https://t.co/0ApitPmDps
#GraphQL Pagination at Meta by Sabrina Wasserman from Meta looks to be an excellent talk for day 1 at #graphqlconf make sure to sign up and get your $200 discount on your ticket with the discount code API
https://t.co/14hJXi1NBp
Omg, #graphqlconf is next week 😊 ...
Get a ticket and use the discount code SPK for a 20% discount!
Here is my personal session schedule:
- @gethackteam: GraphQL in the Era of React Server Components
- @Pascal_Senn: How to Not Break Your GraphQL Clients
- Mauricio Montalvo: Incrementally Adopting GraphQL and Relay at Pinterest
- @Benjie: Techniques to Protect Your GraphQL API
- Andrew Doyle: GraphQL in the House
- @rob_richard: State of defer and stream
- @captbaritone: Semantic Nullability: A Path Toward Safe Non-Null Fields
- @andimarek: What if ... How to Achieve GraphQL Domination
- Pooja Mistry: Not Your Regular Rate Limiting #GraphQL
- @YassinEldeeb7: Workshop: Efficient Cross-Platform GraphQL and State Management with React Native
- @UriGoldshtein: Comparing API Protocols - One Feature at a Time
@apollographql has worked with us in the open for the last year to create an open standard for #GraphQL Federation that lives under the @graphql foundation. I am pretty psyched to talk about all the new stuff that is coming and how it will allow different vendors to interconnect with tooling to enrich customer experience
at #graphqlconf.
https://t.co/16MnBqHIlv
#GraphQLconf is around the corner and the agenda looks amazing. If you want to join us use the discount code SPK to get 20% off your tickets. Hope to meet you in San Francisco!
https://t.co/9p1F63C9fD
🔔 #GraphQLConf Keynote: @michael_staib of @Chilli_Cream will reveal how the community is standardizing #GraphQL as an orchestrator for distributed systems, showcasing new specs, composite schemas, & prototypes: https://t.co/L410UynRrK. Register now 👉 https://t.co/fdOamuJNcb
@Cheesebaron@fboucheros You would have a single client ... but in the deploy pipeline operations are stripped and uploaded to the client registry. This is how graphql is meant to be used and all major GraphQL users like Meta are using it in this way.
@Cheesebaron@fboucheros For this you use a schema registry. Strawberry shake also has support for persisted queries which makes your setup more secure and lets you track what exactly is used by which client in which version. If you do graphql right you have the best api evolution story of any apitech.