Your Agent Is Mine: Measuring Malicious Intermediary Attacks on the LLM Supply Chain
2026-04-09 • Cryptography and Security
Cryptography and Security
AI summaryⓘ
The authors studied how third-party routers that connect large language models (LLMs) to different providers can be risky. These routers see all the data passing through and might inject harmful code or steal secrets because there’s no strong security between clients and providers. By testing many routers, the authors found several that acted maliciously, including some that stole credentials or drained cryptocurrency. They also created a tool called Mine to simulate these attacks and tested ways to defend against them on the client side.
large language modelsAPI routerpayload injectionsecret exfiltrationcryptographic integrityautonomous agentstransparency logginganomaly detectionproxy securitycredential theft
Authors
Hanzhi Liu, Chaofan Shou, Hongbo Wen, Yanju Chen, Ryan Jingyang Fang, Yu Feng
Abstract
Large language model (LLM) agents increasingly rely on third-party API routers to dispatch tool-calling requests across multiple upstream providers. These routers operate as application-layer proxies with full plaintext access to every in-flight JSON payload, yet no provider enforces cryptographic integrity between client and upstream model. We present the first systematic study of this attack surface. We formalize a threat model for malicious LLM API routers and define two core attack classes, payload injection (AC-1) and secret exfiltration (AC-2), together with two adaptive evasion variants: dependency-targeted injection (AC-1.a) and conditional delivery (AC-1.b). Across 28 paid routers purchased from Taobao, Xianyu, and Shopify-hosted storefronts and 400 free routers collected from public communities, we find 1 paid and 8 free routers actively injecting malicious code, 2 deploying adaptive evasion triggers, 17 touching researcher-owned AWS canary credentials, and 1 draining ETH from a researcher-owned private key. Two poisoning studies further show that ostensibly benign routers can be pulled into the same attack surface: a leaked OpenAI key generates 100M GPT-5.4 tokens and more than seven Codex sessions, while weakly configured decoys yield 2B billed tokens, 99 credentials across 440 Codex sessions, and 401 sessions already running in autonomous YOLO mode. We build Mine, a research proxy that implements all four attack classes against four public agent frameworks, and use it to evaluate three deployable client-side defenses: a fail-closed policy gate, response-side anomaly screening, and append-only transparency logging.