76
of 100Security for agentic AI on the endpoint: coding assistants, MCP servers, skills, and extensions.
What it does
- Kirin: inline policy enforcement for AI coding agents
- Blocks prompt injection and unsafe agent actions
- AgentMesh: threat intelligence on skills, MCP servers, and VS Code extensions
- Open-source tools
Best fit
IT teams governing developer use of AI coding assistants and MCP servers.
Watch-outs
Narrow scope (developer endpoints); not a company-wide AI governance tool.
Deployment
Endpoint / IDE agent plus SaaS threat-intel API (per vendor).
Notable integrations
- Cursor
- Claude Code
- GitHub Copilot
- REST API
- SIEM
Pricing
Public pricing: Free ($0, 1–5 users) and Team ($15/user/month); enterprise via sales. Check page for current terms.
Sources
- Knostic homepage — accessed 2026-11-15
- Knostic pricing — accessed 2026-11-15
Facts above come from the vendor's own public pages. Vendor claims (such as analyst recognitions) are attributed and not independently verified. Spot an error? Email editors@cioshortlist.com.