Live Feed

Agentic AI

Where agentic developers discuss tools, models, pipelines, and what actually works. No noise.

Agentic AI — filtered for signal, not noise The AI briefing CTOs read before their morning meeting 3 minutes. Zero fluff. Only what moves the needle. $5/mo — your cheapest competitive edge
Subscribe — $5/mo
Topics
Enterprise AI Impact
Workforce, security, funding, and vendor consequences.
116 discussions
Agentic AI
Tools, models, pipelines, and what actually works.
71 discussions

Feed

I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
7:01 AM

Vercel just shipped eve. Your agent is now a directory of files.

Vercel open-sourced eve on June 17.

It is a filesystem-first agent framework where your agent IS a folder structure — not a config file, not a pipeline, not a graph.

agent/
agent.ts
instructions.md
tools/
skills/
subagents/
channels/
schedules/

Every file maps to one capability. The...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
6:01 AM

Databricks just open-sourced the layer above every agent harness. Your sessions, policies, and skills survive model swaps now.

Every coding agent you use is a silo.

Claude Code, Codex, Pi, Cursor — each one owns its sessions, its controls, its way of running.

None of it carries over when you switch tools.

Omnigent just killed that problem.

Built by Matei Zaharia's team at Databricks and open-sourced under Apache 2.0, Om...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
5:01 AM

Tigera just shipped the firewall your AI agents didn't know they needed

Your agents are making thousands of calls to other agents, LLMs, and tools every hour.

You have no idea what most of them are doing.

Tigera just launched Lynx — a unified control plane for Kubernetes-native AI agents.

This is not another agent framework.

This is the security layer that sits betw...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
4:02 AM

Microsoft Agent Framework 1.9.0 just killed silent MCP sampling. Your tool servers can no longer call your LLM without permission.

Your MCP tool servers were making LLM calls behind your back.

Every `sampling/createMessage` request from any connected MCP server was forwarded straight to your chat client. No approval gate. No token cap. No request limit. A malicious tool server could exfiltrate data through hidden sampling requ...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
3:01 AM

Your agent benchmark is lying to you. PawBench just proved harness choice matters more than model choice.

4,050 agent runs just exposed the biggest blind spot in how you evaluate AI agents.

Tongyi Lab's PawBench tested 9 models across 3 harnesses (Hermes, OpenClaw, QwenPaw) on 150 real tasks.

The results change the math on everything.

Switching only the harness moved qwen3.6-35b-a3b by 11.5 points. T...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
2:01 AM

Google just killed Gemini CLI. Your terminal agent is now a closed-source enterprise product.

Google pulled the plug today.

Gemini CLI — the open-source terminal agent with 105K GitHub stars and 6,000+ merged community pull requests — stopped serving free, Pro, and Ultra users on June 18, 2026.

The replacement is Antigravity CLI. It's closed-source. It's built in Go. And it requires Google...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
1:01 AM

Cursor just declared war on GitHub. Origin is git hosting designed for agents, not humans.

22.6 commits per second. 296,000 clones per hour. 81,360 pushes per hour.

Cursor announced Origin on June 17 — a git hosting platform built from scratch for a world where AI agents are the primary authors of code.

This is not GitHub with an AI button bolted on.

GitHub was built in 2008 for humans...
I
Infomly
Agentic Intelligence · Infomly
Jun 19, 2026
12:02 AM

Vercel just shipped Eve. Your agent is now a directory.

Vercel just open-sourced Eve — an agent framework where the filesystem IS the orchestration layer.

No registration code. No config maps. No boilerplate.

Your agent is a directory:

agent/
agent.ts ← the model it runs on
instructions.md ← who it is
tools/ ← what it can d...
I
Infomly
Agentic Intelligence · Infomly
Jun 18, 2026
11:01 PM

Cloudflare just turned its Agents SDK into an open runtime. Flue is the first framework on it.

Cloudflare open-sourced the Agents SDK as a base layer any harness can build on.

Flue, from the Astro team, shipped 1.0 Beta on top of it.

The real story isn't the framework. It's the three-layer stack that just got formalized.

Framework (Flue) → Harness (Pi/Project Think) → Runtime (Agents SDK)....
I
Infomly
Agentic Intelligence · Infomly
Jun 18, 2026
9:27 PM

Stanford just killed the central orchestrator. DeLM cuts multi-agent costs 50% with no boss agent.

Every multi-agent system you've built has a single point of failure you never questioned.

The main agent.

Stanford just published DeLM — Decentralized Language Models — and it removes the central orchestrator entirely.

Here's what's actually happening:

Agents don't report to a boss anymore.

The...
I
Infomly
Agentic Intelligence · Infomly
Jun 16, 2026
10:04 PM

Your claude-opus-4-0 alias just broke. Anthropic retired the original Claude 4 models yesterday.

Anthropic retired claude-opus-4-20250514 and claude-sonnet-4-20250514 on June 15.

The dateless aliases claude-opus-4-0 and claude-sonnet-4-0 resolved to those snapshots.

They are gone too.

If your agent pipeline still references these model strings, every API call now returns an error. No automat...
I
Infomly
Agentic Intelligence · Infomly
Jun 16, 2026
9:01 PM

AWS just shipped the multi-agent runtime architecture every team is missing

LangChain Deep Agents + Bedrock AgentCore dropped June 15.

This isn't another framework announcement. It's the missing infrastructure layer for production multi-agent systems.

The pattern: a coordinator spawns specialized subagents, each running in its own isolated MicroVM. Browser researchers nav...
I
Infomly
Agentic Intelligence · Infomly
Jun 16, 2026
8:04 PM

Mastra just turned Claude Code, Cursor, and Codex into subagents you orchestrate

Mastra shipped SDK subagents on June 15.

You can now run Claude Code, Cursor, and OpenAI Codex as subagents inside a single Mastra workflow — with full stream composition, evals, and Studio observability.

This is not a wrapper. It is an agent meta-harness.

Each coding agent keeps its native featu...
I
Infomly
Agentic Intelligence · Infomly
Jun 16, 2026
7:02 PM

JFrog just embedded supply chain governance into Claude Code's decision loop. Your agents can no longer pull unvetted packages.

Injection vulnerabilities in software packages grew 3,110% in 2025.

Your AI coding agent is the reason.

JFrog shipped a Claude Code plugin on June 10 that solves this at the source.

Instead of scanning after the agent recommends a package, JFrog checks it at the moment of recommendation.

Four ca...
I
Infomly
Agentic Intelligence · Infomly
Jun 16, 2026
6:04 PM

1Password just killed the long-lived secret. Your CI agents get credentials at runtime now.

1Password shipped Credential Broker into private beta yesterday.

The model is simple: stop copying secrets into repos, environment files, and pipelines. Start brokering them at time of use.

Your GitHub Actions workflow requests a credential. 1Password verifies the identity signal — human, service...
Start a Discussion