Security for the Agentic Era

The arms race is here. We're building the most comprehensive adversarial dataset in AI security. A context-aware, continuously evolving attack vector-database, backing a single API — try it today.

Detect API

Single-call prompt injection detection. Sub-10ms latency, batch support, context-aware scoring.

Quickstart →

MCP Server

Works with Claude Desktop, Cursor, and any MCP-compatible tool. Install via npm or PyPI.

Setup →

Framework Recipes

Ready-made examples for LangChain, LlamaIndex, Vercel AI SDK, Pydantic AI, Mastra, and more.

Browse →

Self-Hostable

MIT licensed. Run on your own infrastructure with Docker Compose. No vendor lock-in.

GitHub →