AnthropicSep 29, 2026
An open model wrote a Chrome exploit in 8 hours
An open-weight model turned a public Chrome patch into a working exploit in 8 hours, for $20.40.
Anthropic research





Automate workflows, investigate vulnerabilities, track down owners, and remediate fast — at scale, with zero engineering friction.
AI turns a CVE into a working exploit in hours, and anyone can use it. Are you ready for the vulns of the future?
Solution
Rival’s Agent Harness gives you the capabilities and context to run security agents at your scale and complexity.
Extend
Partner
Grow
Agentic Toolbox
A few of the agentic super-powers Rival grants your team on day one.
64+
more agents, ready to slot into your security program. Mix, match and run them alongside your team.
Under the hood
Every agent runs on Rival’s Agent Harness: it reads the context of your cloud, code and scanners, and hands the work to specialist subagents.
Every agent action is traced, auditable and held to your policies, from the first prompt to the last change in production.
Agents take security work off your team's plate, from triage to 0-day response and remediation, the moment something happens. Use ours, build your own, or have Rival build them for you.
Every agent works like a seasoned security expert, drawing on specialists built for the hardest parts of the job.
Agents run safely at any scale, with everything they need built in. Keep sensitive work inside your walls by running it on-prem, in your own environment.
Your cloud, code and infrastructure connected in one place, so every agent sees the full picture and remembers what it learned. Built for the largest environments.
We’re on
Build anything with Rival, or pick a built-in agent to start a workflow.
Built-in agents

