How We Built a Stateless ASPM Proxy to Intercept Prompt Injections in <2ms (Zero SDK Friction)

Description and use

SignalGate is a stateless proxy that intercepts prompt injections and token loops with <2 ms latency, requires only a base URL swap, and scales horizontally without SDK friction.

Problem

Shipping LLM-powered applications often requires bulky client libraries or heavy third‑party SDKs, adding latency and complexity, while client‑side guardrails fail to catch sophisticated prompt injections and runaway token loops.

Outcome

SignalGate achieves sub‑2ms latency overhead, zero SDK friction, and stateless scalability, allowing secure LLM integration without codebase bloat.

Workflow

• 1. Point the OpenAI/Claude client base URL to the SignalGate endpoint (base URL swap).
• 2. SignalGate receives the request, inspects the payload using regex/heuristic scanning.
• 3. If a prompt injection or infinite loop is detected, SignalGate blocks the request.
• 4. If safe, SignalGate forwards the request to the LLM provider.
• 5. The LLM provider returns the response to SignalGate, which passes it back to the application.

Evidence

Shipping LLM-powered applications and autonomous agents to production usually forces a tough architectural compromise. You either build custom wrapper logic that bloats your codebase, or you layer on heavy third-party SDKs that add latency, introduce dependency conflicts, and complicate your core routing.

Source

We built SignalGate, a stateless Agent Safety & Performance Middleware (ASPM) proxy that sits cleanly between your application code and your LLM provider (OpenAI, Anthropic, Gemini, etc.). Because it acts as an intelligent proxy layer, it intercepts payloads in flight to evaluate safety, detect prompt injections, and prevent infinite token-loop burns before the request ever hits the model.

Source

Sub-2ms Latency Overhead: Security proxies are useless if they slow down response times. By keeping the proxy stateless and optimizing the regex/heuristic scanning pipelines, the latency overhead sits comfortably under 2ms.

Source

Zero SDK Friction: Developers shouldn't have to rewrite their codebase or learn a new wrapper API. Integration requires just a base URL swap—point your existing OpenAI or Claude client base URL to your SignalGate endpoint, and you're live.

Source

Stateless Scale: No local database overhead or heavy state persistence on the proxy nodes, allowing it to scale horizontally seamlessly behind any standard load balancer.

Source