October 13, 2026 | Virtual
Save my spot
Platform
Solutions
Pricing
Events
Connectors
Resources
Book a demo
Akto Open Source
Akto Cloud
Akto Self-hosted
AktoGPT
Financial services
SaaS
Healthcare
Public sector
E-Commerce
Blog
Academy
DevSecOps
Docs
Developer tools
Community
API CVE database
Home
/
AI supply chain risk now extends beyond models and code to MCP servers, skills, plugins, and agent configurations. Learn how to control artifact, instruction, and execution trust across the full chain.
Krishanu
3 Minutes
Agentic AI Security
Founding Team
Product updates
Research & Insights
News
Vulnerabilities
Announcements
Engineering
Explore AI agent discovery, inventory, A2A and MCP protocols, agent capabilities, permissions, and security controls for managing enterprise AI agents.
4 Minutes
Bhagyashree
Why AI governance policies fail without enforced guardrails - the ownership gap between Legal, Security, and Compliance, and how to close it.
Rushali
How indirect prompt injection reaches AI agents through tool call results - the attack mechanics, real techniques, and the defenses that actually hold up.
Why AI agent skills are the new software supply chain - 1,184 malicious skills found in one campaign, and why static scanning alone can't catch them.
Arpashree
Claude Code, Cursor, Copilot, and Codex security compared - permission models, named CVEs, and the credential-targeting pattern behind every real breach.
How memory poisoning works in AI agents - why it persists across sessions, real 2026 incidents, and the architecture that actually defends against it.
