Back to Home

Blog

Insights on AI agent governance, risk analysis, and shipping faster — safely.

VisionAI AgentsGovernance

Why AI Agents Need Governance — And How to Build It

AI coding agents are shipping code faster than ever, but without governance, speed becomes a liability. Learn why every team using AI agents needs a risk analysis layer and how to build one.

March 6, 20268 min
Read article
GuideGetting StartedTutorial

Getting Started with MergeShield in 5 Minutes

A step-by-step guide to installing MergeShield, running your first risk analysis, understanding scores, and configuring auto-merge for your GitHub repositories.

March 6, 20265 min
Read article
ComparisonEngineeringProductivity

MergeShield vs Manual Code Review: What Changes

How does MergeShield change your team's code review workflow? A practical comparison of automated risk analysis versus traditional manual review — and why the best approach combines both.

March 6, 20267 min
Read article
AI AgentsGovernanceEngineering

Anthropic Says Use More Agents to Fix Agent Code. Here's What's Missing.

Anthropic's multi-agent harness uses a generator and evaluator to improve code quality. But when both agents share the same model, they share the same blind spots. External evaluation changes everything.

March 30, 20267 min
Read article
AI AgentsGovernanceEngineering

What Claude Code's Leaked Source Reveals About AI Agent Governance

Anthropic accidentally leaked Claude Code's full source. The code itself matters less than the unreleased feature flags: autonomous daemons, multi-agent coordination, and a stealth mode that strips AI attribution. Here's what it means for governance.

March 31, 20266 min
Read article
AI AgentsGovernanceEngineering

The Three Signals Undercover Mode Kills (and What to Use Instead)

Undercover Mode strips the three signals most teams rely on to detect AI-generated code. Here's what actually works when attribution is gone.

April 1, 20265 min
Read article
AI AgentsGovernanceEngineering

How a Cursor Agent Deleted 37GB - A Forensic Breakdown

A Cursor agent wiped 37GB by bypassing OS security policies. The forensic breakdown reveals four failure points every team using AI agents needs to fix.

April 2, 20267 min
Read article

Follow us for updates on AI agent governance