Agentic AI Foundation

Agentic AI Foundation

@aaif.io

AAIF is a neutral, open foundation ensuring AI capabilities evolve transparently and collaboratively, advancing adoption of leading open source AI projects.

294 followers2 following553 posts

Your team knows how its work should be done. Agent Skills gives people a way to share that knowledge with AI agents and each other. Our latest blog looks at how teams are building shared skill libraries and keeping their expertise useful as their tools change. Read here: https://bit.ly/4zvMzN3

Still thinking about everything that went down at AGNTCon + MCPCon NA last week? From big announcements to deep dives into MCP and the open agent stack, Day 1 gave us plenty to unpack. Missed it? Or want to revisit your favorite sessions? The full Day 1 replay is ready: https://bit.ly/3U6tMrS

AI infrastructure is hitting physical limits: power, cooling, rack density, and hardware supply chains. Zachary Smith of Datum discusses the constraints and how data centers could reuse waste heat.

An AI agent recommended hiring another engineer. Applying the Google SRE handbook’s lens on operational toil changed the answer. Tim O’Reilly on domain expertise, context engineering, and building more reliable agents.

That’s a wrap on AGNTCon + MCPCon NA! Nothing beats bringing this community together IRL. Thanks to everyone who made these two days so special. As Angie Jones said, once the Christmas tree comes down, 2027 will be here before you know it. So get those travel plans started: https://bit.ly/4yzm7AB

What happens when a goose agent crashes halfway through a task? With the new open source integration between goose and Dapr, it can resume after a restart, preserving its conversation history without repeating completed turns. Explore the integration and examples: https://bit.ly/4dWAWG9

Rao Surapaneni from Google Cloud is breaking down how open standards, observability, and runtime controls enable safer agent deployment. From A2A interoperability to agent governance, this is the infrastructure conversation that matters. Watch live: bit.ly/4dVkjdO

Shadaj Laddad from AWS is tackling exactly that at AGNTCon + MCPCon NA with “Predictable Correctness for Unpredictable Code.” The goal? Use Hydro and Rust to catch distributed systems errors at compile time, even when agents write the code. Catch the keynote live: bit.ly/4dVkjdO

AI-generated code needs verification before it reaches a PR. Erica Hughberg and Demetrios Brinkmann discuss QA sub-agents, workflow-based tests, and using screenshots to catch broken behavior before human review.

Git was built for developers. What does it look like when AI agents start writing the code? Thomas Dohmke, former GitHub CEO and now co-founder and CEO of Entire, is taking the AGNTCon + MCPCon NA stage with “Building the Git Forge of the Future.” Live now: https://bit.ly/4dVkjdO

Sometimes the most powerful AI stories start with love. Paul Conyngham, Founder of Gamgee Technologies, is on stage NOW at AGNTCon + MCPCon NA, sharing how his dog Rosie's cancer journey inspired him to explore a personalised mRNA vaccine with AI. Watch live: https://bit.ly/4dVkjdO

Bigger models? Cool. Better systems? Even better. Jonathan Bryce of the Linux Foundation hit the AGNTCon + MCPCon NA keynote stage to talk about accelerating open intelligence beyond model scaling. Day 2 is LIVE. Tune in: https://bit.ly/4dVkjdO

Congratulations to the first AAIF Annual Technical Excellence Awards recipients! Announced at AGNTCon + MCPCon in San Jose, the awards recognize technical contributions, community building, and innovation with open standards. Meet the winners: aaif.io/blog/winners...

aaif.ioFirst AAIF Annual Technical Excellence Awards Winners - Agentic AI Foundation (AAIF)AAIF announces winners of its first Annual Technical Excellence Awards at AGNTCon, recognizing contributions to MCP, agentic AI open standards, and community building.

Sarah Drasner, Distinguished Engineer at Google, just took the AGNTCon + MCPCon NA keynote stage to talk about “The Agentic Web.” What happens when AI agents become part of how we navigate and interact with the web? This is the conversation you don't want to miss. https://bit.ly/3U6tMrS

Abhishek Choudhary, co-founder and CTO of TrueFoundry, took the AGNTCon + MCPCon NA keynote stage with “From Black Box to Glass Box,” sharing a five-step blueprint for building trustworthy agents. Catch the livestream: https://bit.ly/3U6tMrS

Dawn Song, Professor at UC Berkeley and VP of AI Research at Meta, takes the AGNTCon + MCPCon NA keynote stage to tackle one of agentic AI's biggest challenges: building agents that are both safe and secure. Join the livestream to learn more: https://bit.ly/3U6tMrS

Need a lunch break? Your brain's been running inference all morning. Take five, hit up our game stations, and meet the AAIF robot. One problem: our robot still needs a name. Drop your best suggestions in the comments.

The Solution Showcase is OPEN at AGNTCon + MCPCon North America. Come nerd out over AI agents, MCP servers, integrations, and the technical chaos of making it all work in the real world. The builders are here. Come see what they've been cooking.

Thomas Sarlandie from OpenAI shared a key takeaway during his “MCP in the Wild” keynote at AGNTCon + MCPCon NA: Make every tool discoverable and ensure annotations accurately describe its capabilities. Livestream: https://bit.ly/3U6tMrS

Okay but why are these AI agents actually so cute? Bradley Axen from Block, creator and maintainer of goose, took the AGNTCon + MCPCon NA stage to show us the human-agent workplace. Agents are signed by their humans and inherit their capabilities. Livestream: https://bit.ly/3U6tMrS

Agent Skills is officially joining the Agentic AI Foundation. David Soria Parra from Anthropic took the AGNTCon + MCPCon NA keynote stage for “MCP and the Era of Connectivity.” And that's just one of FOUR new projects joining AAIF today. Get the full announcement: https://bit.ly/47EYADp

Manik Surtani, CTO of the Agentic AI Foundation, and J.R. Storment, Executive Director of the Tokenomics Foundation, brought their perspectives to the AGNTCon + MCPCon NA keynote stage. The conversations are just getting started. Catch the livestream: https://bit.ly/3U6tMrS

4 new projects, 305 members, 15 industry sectors. Mazin Gilbert, Executive Director of AAIF, is live at AGNTCon + MCPCon NA talking growth and what’s next for the agentic AI ecosystem. Mazin is live right now. Don’t miss it. Join the livestream: https://bit.ly/3U6tMrS

Welcome to AAIF, Agent Skills, Open Responses, OpenEnv, and OpenSandbox! These 4 projects support how developers share agent expertise, work across model providers, train agents, and run their tasks in isolated environments. Explore the projects + find out how to contribute: https://bit.ly/4yXyLdG

AI coding ROI isn't measured in lines of code. Pepijn Senders shares how to track token spend against product value and why accepting code you don't understand can create risks in production.

DE LA SOUL and the AI community in one room. Tonight’s AI Community Bash was the perfect kickoff to AGNTCon + MCPCon NA. With the event tomorrow, a big thank-you to Akamai, Anthropic, Auth0, Amazon Web Services, Buzz, Google Cloud, OpenAI, Solo.io, Tokenomics Foundation, TrueFoundry and more.

DE LA SOUL live was already a reason to be here. Then came the raffles. A Nintendo Switch, gift cards, and a chance to meet DE LA SOUL. The AGNTCon + MCPCon and PyTorch communities are having a night. And it’s still happening. AI Community Bash, Hall 3, San Jose Convention Center. Until 9 PM.

The momentum isn't slowing down! Fresh off the keynote stage at PyTorch Conference, Agentic AI Foundation Executive Director Mazin Gilbert is gearing up for AGNTCon + MCPCon North America! See you there! Still not too late to register: https://bit.ly/4vkntyD

AI can write your code, but who's cleaning up the mess? Dexter Horthy breaks down why passing tests doesn't mean maintainable code. Catch him at AGNTCon + MCPCon North America, Oct 22–23 in San Jose. It's not too late to register: https://bit.ly/4lT6ZJM

AI agents can test thousands of attack paths in parallel. A small network vulnerability in a supposedly isolated sandbox can become an escape route. Erica Hughberg and Demetrios Brinkmann discuss agent security and bots with buying power.

AWS's James Ward goes beyond “hello world” to tackle real-world MCP challenges, from observability to smarter tool design. Catch him at AGNTCon + MCPCon NA as he shares practical patterns for turning basic MCP APIs into robust agent integration interfaces. Register: https://bit.ly/4vkntyD

Have you seen the AGNTCon + MCPCon NA memes making the rounds? 👀 Our agents saw them and decided they wanted in on the fun. Going to AGNTCon + MCPCon North America? Create your own, tag us, and show us who's joining the crew in San Jose, October 22–23! 🔗 https://bit.ly/4vkntyD

Agents need narrow permissions, protocols that work together, and costs measured per completed task. At AGNTCon + MCPDev Summit EU, maintainers discussed what it takes to build and operate agentic systems. Read the recap: aaif.io/blog/maintai...

Browser agents scraping DOMs and screenshots are slow, token-intensive, and unpredictable. Sarah Drasner of Google explains how WebMCP exposes defined actions, flags consequential operations, and helps developers guide agents safely.