Cynative
Open-source read-only security agent framework for cloud infrastructure and code platforms
About Cynative
Cynative is an open-source security agent framework that provides real-time, read-only infrastructure access, connecting AWS, GCP, Azure, EKS/GKE/AKS, self-managed Kubernetes, GitHub, and GitLab to help check privilege-escalation, public-exposure, and credential-leak risks. It connects to multiple LLMs through the embedded Bifrost SDK and uses terminal credentials for tool calls with interactive approval.
Best for
Best use cases
Check cloud and Kubernetes privilege-escalation risks
Investigate publicly exposed infrastructure
Audit GitHub and GitLab connections
Check for credential leaks
Features
Multi-cloud and code-platform connections
Connect AWS, GCP, Azure, EKS/GKE/AKS, self-managed Kubernetes, GitHub, and GitLab.
Secure read-only access
Use credentials already available in the shell to access infrastructure, with least-privilege read-only credentials recommended.
Multiple LLM providers
The embedded Bifrost SDK supports OpenAI, Anthropic, Azure OpenAI, Amazon Bedrock, Google Vertex/Gemini, Cohere, Mistral, Groq, Ollama, vLLM, and more.
Interactive approval
Approve tool calls individually: y runs once, a allows later calls to that tool in the session, and other keys reject.
Pros
- Covers multi-cloud, Kubernetes, GitHub, and GitLab
- Real-time read-only infrastructure access
- Supports multiple LLM providers
- Supports per-call tool approval