- Tool type
- Security layer for AI coding agents and IDEs
- Made for
- Developers, engineering teams, and security leads
- Runs on
- Cursor, Claude Code, Copilot, Windsurf on macOS and Windows
- Linux in beta; deployment via endpoint client and IDE extension.
What it does
Kirin watches what AI coding assistants do as they work: prompts, tool calls, MCP server connections, extensions, package installs, and instruction files such as CLAUDE.md or .cursorrules. It checks that activity against configured policies in real time and produces alerts, blocked actions, risk scores, dashboards, and audit logs.
Developers and security teams use it to catch secrets pasted into chat, destructive commands, prompt injection hidden in MCP tool descriptions, hallucinated or malicious packages, and unapproved MCP servers before the agent acts. Policies are enforced inline, and an endpoint client also surfaces shadow AI use on the machine.
It suits engineering teams and security leads using Cursor, Claude Code, GitHub Copilot, Windsurf, or similar agents. A Free plan covers individuals and teams of up to five users, while Team adds a unified dashboard, role-based access control, and audit logs. Deployment uses an endpoint client and IDE extension, with macOS and Windows supported and Linux in beta.
What it can do
- Blocks secrets in chat content before they are transmitted
- Flags destructive agent commands such as rm -rf or table drops
- Detects prompt injection in MCP server tool descriptions
- Scans packages installed by coding agents for malware and vulnerabilities
- Monitors MCP servers, IDE extensions, and plugins against an allow list
- Provides dashboards, alerts, and audit logs for security teams
Works with
- Cursor
- Claude Code
- GitHub Copilot
- Windsurf
- Codex
- JetBrains
- Gemini CLI
- SIEM
- Slack
Before you choose
- Kirin enforces at the agent layer; Shadow AI Discovery is visibility-only.
- Some agents such as Claude Code and Cowork need the endpoint client plus agent hooks.
- Advanced contextual detection is an add-on on Team and included in Enterprise.
Listed under
Get started
Free plan for individuals and teams up to five users.
- Create a Kirin console account and choose the Free or Team plan.
- Install the Kirin endpoint client on developer machines, directly or through your MDM.
- Install the IDE extension for your agent and start in Monitor mode before switching to Enforce.
Plans and prices
Prices in USDTeam is $15 per user per month for 5+ users. Enterprise pricing is custom via contact. The 14-day trial applies to Team and Enterprise.
- Continuous scanning across projects
- MCP scanning
- Unified dashboard
- Audit logs
- Enterprise SSO
- Compliance Ready Reporting
14-day free trial.
Source: knostic.ai/pricing, checked 29 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
- Kirin — Security for AI Coding Agents - Knostic ↗www.knostic.ai
- Pricing - Knostic ↗www.knostic.ai
- Knostic - The control plane for agentic AI on the endpoint ↗www.knostic.ai
- AgentMesh — AI Agent Supply Chain Security - Knostic ↗www.knostic.ai
Ratings and reviews
No ratings for Knostic Kirin yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- knostic.ai: product, pricing and policy pages
- Web search
- Confirms the official site
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
- manual_review
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.