---
title: "Snyk launches Evo Continuous Offensive Security with AI pentesting"
slug: "snyk-launches-evo-continuous-offensive-security-with-ai-pentesting"
published: "2026-08-12"
beat: "Launches"
tags: ["Launches"]
creator: "Agentry Newsroom"
editor: "Susanne Sperling, Editor — Human in the Loop"
tools: ["Claude (Anthropic)", "Perplexity Sonar"]
creativeWorkStatus: "verified"
dateReviewed: "2026-08-12"
aiActArticle50: "compliant"
humanView: "https://agentry.news/launches/snyk-launches-evo-continuous-offensive-security-with-ai-pentesting"
agentView: "https://agentry.news/agent/snyk-launches-evo-continuous-offensive-security-with-ai-pentesting"
---# Snyk launches Evo Continuous Offensive Security with AI pentesting

> Snyk announced general availability of Evo Continuous Offensive Security on August 4, 2026, an autonomous AI-powered pentesting platform that continuously attacks applications as they change and deliv

*Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. [AI policy](/ai-policy).*

Snyk announced general availability of **Evo Continuous Offensive Security (Evo COS)** on August 4, 2026, positioning the platform as an **autonomous, AI-powered pentesting and agent red teaming** system that continuously probes applications for exploitable vulnerabilities [Snyk](https://snyk.io/news/snyk-launches-evo-continuous-offensive-security/). The release marks a shift in how organizations approach offensive security testing—moving from periodic manual penetration tests to continuous, agent-driven vulnerability discovery.

## Autonomous Pentesting at Scale

Evo COS operates by autonomously attacking applications as they change, according to [SecurityBoulevard](https://securityboulevard.com/2026/08/snyk-brings-evo-continuous-offensive-security-to-general-availability-at-black-hat/). Rather than waiting for scheduled security assessments, the platform deploys AI agents to identify and validate exploitable weaknesses in real time. Snyk described the product as delivering **"validated proof of exploitability,"** meaning the agents don't merely flag potential issues—they demonstrate concrete attack paths that security teams can act on immediately.

## Agent Red Teaming Module

**Agent Red Teaming**, delivered as part of Evo COS, launched alongside the core platform [BusinessWire](https://www.businesswire.com/news/home/20260804713463/en/). This capability allows organizations to simulate adversarial AI-powered attacks against their own systems, helping teams understand how malicious agents might exploit their applications. The dual release underscores Snyk's strategy: as AI agents become more prevalent in attack vectors, defensive infrastructure must evolve to match that sophistication.

## Market Context

Snyk, which describes itself as **"the AI security company,"** positioned Evo COS amid growing concern about AI-driven attack surfaces. The announcement came at Black Hat USA 2026, a venue where both offensive and defensive security vendors preview new capabilities to enterprise and government audiences. The timing reflects an industry-wide recognition that traditional vulnerability scanning—designed for static codebases and manual workflows—cannot keep pace with the speed and complexity of modern, AI-augmented applications.

## Immediate Availability

Both Evo Continuous Offensive Security and **Snyk Secrets** became generally available "today," according to Snyk's statement [Snyk](https://snyk.io/news/snyk-launches-evo-continuous-offensive-security/), meaning customers can begin deploying the agents into their environments immediately. The product is not in beta or preview; it is production-ready.

This launch represents a concrete expansion of the AI agent economy into the security operations space—a domain where agent actions directly impact organizational risk posture and compliance outcomes.