Local security audit for AI API relays and LLM proxies: detects prompt injection, model substitution, tool-call rewriting, SSE anomalies, error leakage, and Web3 wallet risks.

Verify which LLM an OpenAI-compatible API really serves — single-token behavioral fingerprinting (Jensen-Shannon) from the paper "One Token Is Enough" (arXiv:2607.10252). Zero-dependency TypeScript library + CLI. Catches model substitution by resellers & gateways.

Which model is really behind your API relay or agent IDE? Behavioral fingerprinting + anytime-valid sequential tests (FPR<=1%). LLMs can't be random - measured on 9 frontier models.

一个用 TEE 远程证明和响应签名实现的用户可验证 AI 中转方案。 A user-verifiable AI relay design implemented with TEE remote attestation and response signing.

CLI framework for auditing LLM providers and detecting model substitution.

Is your LLM provider serving the model you're paying for? A detector with a MEASURED FPR