Anthropic releases 817-strong cybersecurity skills library for AI agents
The Anthropic Cybersecurity Skills project provides a structured knowledge base tailored for AI agents. The library covers 817 skills across 29 security domains, from forensics to cloud infrastructure protection.
Each skill aligns with the agentskills.io standard and maps to six industry frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, MITRE D3FEND, NIST AI RMF, and MITRE Fight Fraud Framework. This enables agents to follow step-by-step action plans akin to a senior analyst’s workflow.
The library integrates with Claude Code, GitHub Copilot, Cursor, and other MCP-compatible platforms. The tool is open-sourced on GitHub under the Apache 2.0 license.
Up next
Claude Code update version 2.1.187
Claude Code 2.1.187 adds key security and stability updates. You can now restrict isolated commands from accessing secrets. MCP tool calls that hang are automatically interrupted after 5 minutes.
Related stories
- Claude Duet won’t share your card number even if you ask
- Claude’s auto mode skips connector “Always ask” settings
- Anthropic’s test agents filled out US visa applications
- Claude desktop teardown: MCP servers run outside the VM
- Backdoored skills bypass Anthropic’s scanner
- Users report Claude attempting to guess passwords
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.