Skip to content
← All news

January 23, 2026

8 curated links on AI agents, Python, security, Google Cloud, and engineering, selected by Médéric Hurier.

  1. Claude’s Constitution

    anthropic.com

    Anthropic has published "Claude’s Constitution," a detailed document outlining the values, safety guidelines, and ethical principles used to train and steer their AI model, Claude.

  2. How Google SREs Use Gemini CLI to Solve Real-World Outages

    cloud.google.com

    Google SREs are using Gemini 3 and the Gemini CLI to automate incident response steps, from classification and mitigation to root cause analysis and post-mortem generation, effectively reducing "Bad Customer Minutes".

  3. Gemini in Chrome is getting “Skills” as it moves toward becoming a full AI agent

    chromeunboxed.com

    Gemini in Chrome is evolving from a chatbot into an active "agent" with a new "Skills" feature that allows users to define custom instructions for handling specific tasks and workflows directly within the browser.

  4. How to use the DevTools MCP server for AI debugging #DevToolsTips

    youtube.com

    A video by Chrome for Developers explaining how to use the DevTools MCP (Model Context Protocol) server for AI debugging.

  5. Personal Intelligence in AI Mode in Search: Help that's uniquely yours

    blog.google

    Google AI Pro and Ultra subscribers can now opt-in to connect Gmail and Google Photos to AI Mode in Search for personalized results and recommendations.

  6. Benefits and Challenges of Spec-Driven Development (and how Antigravity is changing the game)

    medium.com

    An article discussing the pros and cons of specification-driven development and introducing a new tool or concept called "Antigravity" in the context of Google Cloud.

  7. Agent to Agent Protocol (A2A) - AI Agents in a Nut$SHELL #1

    youtube.com

    A video by Médéric Hurier (Fmind) discussing the Agent to Agent (A2A) Protocol, likely part of a series on AI Agents.

  8. Agent to Agent Protocol (A2A) - Bleeding Agent #1

    youtube.com

    Another video by Médéric Hurier (Fmind) demonstrating the Agent to Agent Protocol, likely showing a "Bleeding Agent" implementation.