<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Posts on Twisha Bansal</title><link>https://twishabansal.github.io/posts/</link><description>Recent content in Posts on Twisha Bansal</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 28 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://twishabansal.github.io/posts/index.xml" rel="self" type="application/rss+xml"/><item><title>Your Databases Finally Speak Human</title><link>https://twishabansal.github.io/posts/skills-in-claude-codex/</link><pubDate>Tue, 28 Apr 2026 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/skills-in-claude-codex/</guid><description>Google Cloud launches AI agent skills for 12 data products across Gemini CLI, Claude Code, Codex, and Antigravity, letting you query databases conversationally.</description></item><item><title>Your Gemini CLI Extensions Just Got Smarter: Introducing Agent Skills</title><link>https://twishabansal.github.io/posts/skills-rollout/</link><pubDate>Wed, 18 Mar 2026 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/skills-rollout/</guid><description>Google&amp;rsquo;s Cloud SQL for PostgreSQL Gemini CLI extension now uses Agent Skills for faster, leaner, on-demand capability loading.</description></item><item><title>The Secret Sauce of Reliable AI: Implementing Robust Pre/Post-Processing Hooks</title><link>https://twishabansal.github.io/posts/pre-post-processing/</link><pubDate>Tue, 17 Feb 2026 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/pre-post-processing/</guid><description>Learn how pre/post-processing hooks at the agent, model, and tool levels make AI agents safer, more reliable, and enterprise-ready.</description></item><item><title>From Scripts to Stories: Preserving Developer Flow with MCP Prompts</title><link>https://twishabansal.github.io/posts/prompt-support/</link><pubDate>Tue, 25 Nov 2025 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/prompt-support/</guid><description>MCP Toolbox&amp;rsquo;s new prompt support lets developers define reusable YAML workflows that turn multi-step database investigations into single conversational commands.</description></item><item><title>Chain of Trust for AI: Securing MCP Toolbox Architecture on Cloud Run</title><link>https://twishabansal.github.io/posts/cloud-run-deployment/</link><pubDate>Wed, 03 Sep 2025 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/cloud-run-deployment/</guid><description>Learn to build a zero-trust security model on Cloud Run, securing MCP Toolbox database access through service identities and chained credentials.</description></item><item><title>Seamless Tool Integration for Agents: A Deep Dive into the new Toolbox JS SDK</title><link>https://twishabansal.github.io/posts/js-sdk-announcement/</link><pubDate>Tue, 24 Jun 2025 00:00:00 +0000</pubDate><guid>https://twishabansal.github.io/posts/js-sdk-announcement/</guid><description>The new Toolbox JS SDK simplifies connecting AI agents to databases by turning YAML-defined tools into native JavaScript functions for frameworks like LangChain and Genkit.</description></item></channel></rss>