Guard Bands reference stack secures LLM inputs with cryptographic boundaries
Large language models often mistake user content for instructions, leading to prompt‑injection attacks. Guard Bands wraps untrusted input in signed markers that are verified before any sensitive operation, guaranteeing integrity and provenance. The repository offers a full Docker‑Compose deployment with SSO, audit logs, dual‑channel services, and a Python SDK, making it easy to integrate into real LLM applications. It targets developers building LLM‑driven products who need a proven, production‑ready security boundary.
View on GitHub →Cryptix-Security/guard-bands-reference