Full KeyRank table

Every qualifying story, ranked. How KeyRank works

#	Story	Pillar	KeyRank	Source
1	Anthropic follows OpenAI in admitting its Claude models reached out of test environments and attacked real-world systems	Agents	78	The Decoder
2	Anthropic says Claude accidentally hacked real companies too	Agents	78	The Verge AI
3	Anthropic says its Claude models 'gained unauthorized access' to other organizations' systems	Agents	78	CNBC Technology
4	Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests	Agents	78	Wired AI
5	Anthropic’s Claude AI models hack into 3 outside groups during testing	Agents	78	Financial Times Technology
6	Perplexity Open Sources Numbat To Monitor Risky AI Coding Agents	Agents	72	Forbes Innovation
7	Anthropic says its own AI models breached three companies during security tests	Agents	72	TechCrunch AI
8	Nous Research Ships Three Integration Paths for Hermes Agent and Buzz, Block’s Open Source Nostr Workspace for Humans and Agents	Agents	62	MarkTechPost
9	Icite targets knowledge graphs to give AI agents the context needed for autonomous security workflows	Agents	52	SiliconAngle
10	Optimizing production agents with Amazon Bedrock AgentCore Observability	Agents	48	AWS Machine Learning Blog
11	Building a Policy-Governed Multi-Agent Financial Research Workflow with Omnigent	Agents	45	MarkTechPost