AAIF is a neutral, open foundation ensuring AI capabilities evolve transparently and collaboratively, advancing adoption of leading open-source AI projects.
Joined January 2026
- Tweets 458
- Following 4
- Followers 1,924
- Likes 133
353 Photos and videos
What causes more MCP incidents: malicious actors or missing guardrails?
Cecilia Liu of member company @Docker shares that many failures come from weak operational controls, not protocol flaws.
Her approach: MCP gateways, sandboxed runtimes, and integration with enterprise identity and observability systems.
See how a three-layer control plane can prevent the kinds of failures that only show up in production: bit.ly/4uvVf2i
1
2
141
Most docs are written for humans. What about agents?
Member company Microsoft's Learn MCP Server gives any MCP-compatible agent live access to learn docs across terminals, IDEs, and custom assistants โ a model for making docs and SDKs agent-ready.
Read more: bit.ly/4x2WOYe
2
141
What does it take for coding agents to stay useful over longer runs?
Cursor Ships Composer 2.5, a stronger model tuned for sustained coding work, with signals that an even larger model is already in development. Like Composer 3.0, it defaults to an open model, reinforcing the shift toward more flexible model choices in agentic coding workflows.
From @cursor_ai on X: bit.ly/4vhO9zo
173
๐ A heartfelt thank you to our #MCPDevSummit Mumbai diamond sponsor AWS. It's not too late to join in on the fun 14-15 June in Mumbai! Connect with developers working on secure integrations between AI systems & real-world applications. AWS Keynote Session: sched.co/2P0t7 Register: bit.ly/4vcG4wR Schedule: bit.ly/3RpxT0p
1
140
๐ฅ Just 2 days until #MCPDevSummit Mumbai!
See who's joining developers, founders, architects, & #AI builders from across the ecosystem as the community gathers in Mumbai from 14-15 June.
๐ See who's attending: bit.ly/4vB1ym8
๐๏ธ Register: bit.ly/4vcG4wR
2
132
What gets built next in #AgenticAI will be shaped by the people who share their work now. Last chance to submit to speak at #AGNTCon #MCPCon Japan. Deadline: TONIGHT, June 12, 11:59 PM JST
๐ Tokyo | Sept 10-11
๐ bit.ly/49Q0OB9
6
193
Agentic AI Foundation retweeted
As the MLOps Community becomes @AgenticAIFdn's user group, we're launching Agentic Conversations- a new podcast and video series with the builders, researchers, and practitioners working on agentic AI systems.
2
1
3
301
What if your security team could investigate threats through agents?
Member company @Google open sourced a framework that brings AI agents into Chrome Enterprise security operations โ letting teams investigate threats, automate analysis, and query browser telemetry through agent workflows.
Read more: bit.ly/4o2jdRm
2
139
What does โmemoryโ mean for AI agents beyond a chat history?
TencentDB Agent Memory (@TencentGlobal) introduces a local, MIT-licensed system that combines symbolic short-term memory with a four-tier long-term structure for conversations, atomic facts, scenarios, and personas. A lightweight approach to keeping agent context structured over time.
Explore the details from @MarkTechPost: bit.ly/4u7f8wK
4
2
190
#AGNTCon #MCPCon Japan brings the #AgenticAI community to Tokyo this September.
Tomorrow is your final opportunity to join the speaker lineup.
โฐ CFP closes June 12 at 11:59 PM JST
๐ bit.ly/49Q0OB9
1
6
287
TODAY IS THE DAY! ๐จ The #CallForProposals for #AGNTCon #MCPCon China (Sept 6-7) closes at 11:59 AM CST today. Help us build the future of #AI agents.
Submit NOW: bit.ly/3OuRLhS
118
Free-token era may be ending.
@goose_oss maintainer Douwe argues agent systems will need dynamic routing across closed, open, and local models.
Workflows, memory, and permissions shift into the agent layer as costs and availability change.
Early signal: OpenAI already routes some Codex traffic to open models.
Explore the post by Douwe: bit.ly/4u6hJH9
1
5
11
1,206
What matters more in production: the model or the execution architecture behind it?
Misha Lanin, Head of Growth at Port of Context, shares findings from evaluating agent execution patterns with @goose_oss . The results show how execution architecture can influence token efficiency, cost, and reliability at scale.
The future of agent systems will be shaped by more than model quality alone.
Read more: bit.ly/4ey3pCG
1
4
12
594
Save now on your ticket to #AGNTCon #MCPCon North America! Early bird pricing ends July 28 ๐ bit.ly/4vkntyD
Join us in San Jose October 22-23, where the agentic stack is being built!
1
107
It's a wrap! The energy at #MCPDevSummit Bengaluru was amazing! A sincere thank you to everyone who attended. Up next, Mumbai!
2
2
11
3,387
How do you safely run an AI agent in production?
Anthropic breaks down how it sandboxes Claude: the isolation mechanisms, security controls, and operational safeguards behind running it at scale.
See their playbook: anthropic.com/engineering/hoโฆ
3
6
232