Anthropic Cybersecurity Skills
β Text source-checked Nov 11, 2026Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
In short
An open-source library of 754 cybersecurity skills for AI agents, mapping to MITRE ATT&CK and NIST standards. It deploys to Claude Code, Copilot, and 20+ platforms.
Anthropic Cybersecurity Skills is an open-source library of structured cybersecurity skills that follows the agentskills.io standard. The library deploys to Claude Code, GitHub Copilot, Cursor, and over 20 AI coding platforms.
Each skill references MITRE ATT&CK technique IDs and includes mappings to NIST CSF 2.0, MITRE ATLAS, D3FEND, and NIST AI RMF where applicable. The tool uses a progressive disclosure pattern where AI agents read YAML frontmatter to determine relevance before loading the full workflow.
The collection contains 754 skills across 26 security domains. The domains covered range from cloud security and OT/ICS to red teaming and compliance.
The project is a community effort and is not affiliated with Anthropic PBC.
According to Anthropic Cybersecurity Skills: Anthropic Cybersecurity Skills is an open-source library of structured cybersecurity skills that follows the agentskills.io standard. Its pages list this strength: Contains 754 skills across 26 security domains. They also state this limitation: Installation via npx requires Node.js version 18 or higher.
β The AI Tool Bible editorial team
Pros
- β Contains 754 skills across 26 security domains.
- β Skills are mapped to five security frameworks.
- β Available as open source under the Apache 2.0 license.
- β Uses a progressive disclosure pattern to minimize token usage during skill discovery.
Cons
- β οΈ Installation via npx requires Node.js version 18 or higher.
Facts
- Platforms
- cli (not source-checked)
- Pricing
- free - Open source, Apache 2.0 (not source-checked)
- Open source
- Yes (license: Apache-2.0)
- GitHub stars
- 35,801 (checked Nov 9, 2026) (not source-checked)
- Last GitHub push
- Aug 31, 2026 (not source-checked)
- First commit
- Feb 2026 (not source-checked)
- Earliest Wayback capture
- 2026-03-18 (earliest archived capture of the site, not a launch date)
Where this tool sits
Use cases
Frequently asked
- How much does Anthropic Cybersecurity Skills cost?
- The tool is free and open source under the Apache 2.0 license. It is a community effort and is not affiliated with Anthropic PBC, allowing unrestricted access to the skill library.
- Which AI coding platforms support this library?
- It deploys to Claude Code, GitHub Copilot, Cursor, and over 20 other AI coding platforms. The skills follow the agentskills.io standard, ensuring compatibility across these diverse development environments.
- What security standards do the skills reference?
- Each skill references MITRE ATT&CK technique IDs. Where applicable, they include mappings to NIST CSF 2.0, MITRE ATLAS, D3FEND, and NIST AI RMF, providing structured alignment with major industry frameworks.
- How many skills and domains are included?
- The collection contains 754 skills across 26 security domains. Coverage ranges from cloud security and OT/ICS to red teaming and compliance, offering a broad scope for various security operations.
- How do AI agents use these skills?
- The tool uses a progressive disclosure pattern. AI agents read YAML frontmatter to determine relevance before loading the full workflow, optimizing how the agent processes and applies the specific cybersecurity knowledge.