The StackHawk Blog

Security research, product news, and opinions on what happens when AI writes most of your code.

A dark-themed chart shows a four-step ladder from autocomplete to full auto, highlighting delivery as the key metric and noting wrong metrics like code accuracy, agent calls, and token burn at each rung.

The AI Coding Maturity Ladder, and the One Metric That Survives It

Every engineering org using AI coding agents is climbing the same ladder. Most don't know they're on it. And at...

Read More

What We Learned Watching Engineers Use AI Agents for Real Work

We spent months watching engineers use AI coding agents on real engineering work. Not demos. Not benchmarks. Actual teams, actual...

Read More

AI Coding AI Security Shift Left Security

Finding Went Exponential. Your Fixing Didn’t.

Defenders and threat actors got handed the same superpower this year. It finds vulnerabilities. The CVE curve is the cleanest...

Read More

Diagram comparing Two queues with review and security queues, and One loop with a flowchart of scanning, fixing, verifying, and cleaning PRs, showing a more streamlined process.

AI Coding AI Security

A Better PR Inbox Doesn’t Fix the Queue. It Just Sorts It.

GitHub shipped a redesigned pull request dashboard this week. Inbox, saved views, smarter filters, keyboard shortcuts. The author filter now...

Read More

An Eval Harness for Agent Skills: How We Change Behavior on Purpose

Agent skill experiment dashboard comparing old and new skill versions across three repos with a blind evaluator scorecard

Read More

Interested in seeing StackHawk at work?

Schedule time with our team for a live demo.

Get a Live Demo