Blog Archives

Solving the AI Agent Architecture Gap in Modern Development

Developers building with Claude Code are converging on the same solutions without talking to each other. Nobody coordinated this.

Running multiple sessions: VS Code vs Discobot

If you’ve ever tried vibe coding, or even just asked an AI coding agent to do some sort of work for you, you’ve probably thought to your self: It’s annoyingly boring to wait for this agent to finish its task. So what do you do when one agent isn’t enough, and you maybe want to […]

The Trust Crisis: Why MCP Security Fails Autonomous Agents

The sessions on agentic AI at RSAC 2026 were packed because practitioners were scared, and smart enough to know it.

Simplifying MCP Server Deployment with Python and Obot MCP Gateway

A practical guide to creating, packaging, and hosting an MCP server using Python and Obot’s enterprise AI platform.

Dangerous MCP OAuth Shortcuts are Ruining Security

Hundreds of production servers are already compromised or dangerously exposed, and the organizations running them likely have no visibility into it.

What Uber’s MCP Journey Tells Us About Enterprise MCP Strategy

Learn about Uber's enterprise MCP strategy and its practical applications shared at the MCP Dev Summit North America.

The Claude Leak Crisis: A Masterclass in MCP Security

When a Build Oversight Becomes an Involuntary Open-Source Project Anthropic shipped a source map by accident. By the time anyone noticed, 3,800 developers had already downloaded it, and the MCP security implications were sitting on their machines in 512,000 lines of unobfuscated TypeScript. The instinct is to treat this as an embarrassing stumble by a […]

MCP Enterprise Architecture That Actually Works: The Complete Reference Guide

The final post in Obot's 10-part series on enterprise MCP. Pull together identity, access, audit, and data layers into a single reference architecture — with a checklist to evaluate where you are and what's left to build.

MCP PII Data Security: How Tool Calls Leak PII and How to Stop It

MCP tool calls can silently expose sensitive data. Learn how PII leaks into LLM context and how to prevent it with control plane filtering.

MCP Prompt Injection: Why Your AI Agents Can’t Defend Against It Alone

MCP prompt injection is a structural risk for AI agents. Learn how attackers exploit autonomous tool retrieval and how to defend with layered infrastructure controls.

Claude Code Tips: The Master Guide to Advanced Agent Workflows

The right claude code tips aren't about working harder or writing longer prompts; they're about understanding the mechanics underneath. This guide shows you the real patterns to use.

Navigating MCP Architecture’s Awkward Adolescence

If you’ve been building with MCP for any length of time, you’ve probably already discovered that the protocol works exactly as advertised, and that working as advertised creates its own set of problems. The MCP architecture that solved the N×M integration nightmare has generated a new category of friction: context bloat from over-connected servers, static […]