Control plane for agentic AI security on the endpoint
High-assurance infrastructure for AI agents
Stage
Not given
Not given
Founded
Not given
Not given
Based in
US
US
Pricing model
Freemium
Open source
Pricing
Kirin: Free ($0/mo) for individual developers and small teams (1-5 users), Team plan at $15/mo per user for growing teams (5+ users), Enterprise plan with custom pricing. AgentMesh: Free ($0/mo) with 300 items per category, Premium at $15/mo per user, Pro at $850/mo per user for unlimited API access
Not given
Key features
Policy enforcement before agent execution
Blocks prompt injections and rogue agent actions
Real-time detection of malicious behavior
MCP server scanning and blocking
Shadow AI discovery across organization
Secrets detection and prevention
Skills and extensions vetting
Dangerous artifacts rejection
Support for Cursor, Claude Code, GitHub Copilot, Windsurf
Minute-level deployment
Harpe: chooses what an AI agent reads under a token budget with consistent results
Kibisis: keeps agent memory on your hardware with AES-256-GCM encryption
Argus: append-only record that detects edits with cryptographic verification
No network code by default for local and disconnected use
MIT License with published code, benchmarks and raw results
What makes it different
Visibility into shadow AI across enterprise systems
Unified control plane for multiple coding agents and IDEs
Supply chain threat discovery before components run
Secure coding enforcement with threat modeling
Measured empirical validation across 6,111 questions outperforming full-context models
Zero plaintext on disk with encrypted vaults matching plaintext accuracy
78.3% fewer tokens sent to the model with a 1,200-token budget per turn
Preserves conversational chronology and relational qualifiers under token budgets
Knostic
Perseus Computing