v0.2.2 is live 🚀
https://chromewebstore.google.com/detail/linlkeaipfpnhddjkpcbmldionajfifa?utm_source=item-share-cb

#Agentic #Workflow is now available on the #Chrome Web Store with new improvements and fixes.

You can build workflows directly in your #browser to:
• scrape data
• use #AI (local or not)
• #automate tasks on any webpage

Still early : #feedback really helps us shape it:
https://awflow.io

No other country on Earth has as much faith in #agentic #AI as the US.

Proof:
We installed a Russian agent in the White House, and his IQ tests prove that his intelligence, if any, is artificial.

Why Settle for One AI Assistant When You Can Have Two?

Two weeks ago, I discovered that Mistral.ai also provides a coding assitant, similar to GitHub Copilot (GHC), called Mistral Vibe (GitHub page).

In those two weeks I’ve been using Mistral Vibe in parallel to GHC. Just because I wanted to try and see the difference! And just after a couple of days I noticed that the agents definition in Mistral Vibe are a bit different from GHC (in hindsight: […]

Observer | Why Agentic A.I. Deployments Are Failing Before They Scale by David Stokes

Agentic A.I. is no longer a technology on the horizon. It is being deployed today in live enterprise environments, with real operational consequences. In 2026, the conversation in most boardrooms has already shifted from “should we pay attention to this?” to “how do we move safely and most effectively?”

The vendor landscape is not making these questions easier to answer. Incumbent software companies—the platforms already embedded in enterprise architecture—are racing to layer agentic capabilities onto their existing suites, repositioning products many organizations already own. Simultaneously, a new generation of companies built natively on agentic architectures is entering the market, often targeting the same workflows with different approaches. The result is a market that is genuinely moving fast and generating noise in roughly equal measure.

In that environment, the promotional narrative tends to dominate. Early wins get amplified. Failure cases stay private. The gap between what vendors are projecting and what enterprises are experiencing in deployment is wider than it should be at this stage of the technology’s maturity. Executives are being asked to make significant capital and operating model commitments against a signal-to-noise ratio that is, at best, unfavorable.

The cost structure is real—and so is the return

Risk exposure has new dimensions

The operating model problem

For organizations already in deployment

What determines the outcome

Read more: https://observer.com/2026/04/agentic-ai-operating-model-enterprise-adoption/

#agentic-ai #artificial-intelligence #business #technology #customerservice

#Agentic systems need more than prompts and model calls. They need a surrounding machinery that helps the system observe itself and improve, including logs, metrics, traces, tests, and specifications.

This post [1] gives a good rundown of the key considerations:
1. Generic quality scores like helpfulness or #hallucination are often not very useful because they do not explain what actually failed. Define narrow metrics tied to specific failure modes. (1/4)