Governance  ·  Glossary

Agentic AI Skills and Integrations (MCP Skills)

Atomic tools and capabilities that AI agents can invoke to interact with external systems—such as database queries, API calls, or file operations. Skills are published in open marketplaces and can be poisoned, hijacked, or exploited. The OWASP Agentic Skills Top 10 (AST10) is the first framework for securing them.
Agent skills are the primary attack surface for tool poisoning, prompt injection, and supply-chain compromise; as agents and skills proliferate, unsecured skill marketplaces become vector hubs for enterprise attacks.
References
OWASP - Agentic Skills Top 10 (AST10)
Track this in the live feed See how this plays out in real AI security and governance developments.
Open the feed →