rag-inject-guard is a Python library designed to protect RAG systems from indirect prompt injection. It scans retrieved documents for instruction overrides, system-prompt manipulation, exfiltration attempts, and invisible Unicode or homoglyph smuggling in both English and Turkish. Malicious documents are quarantined before they can influence the model. Released under Apache-2.0, it provides developers with a practical security layer for retrieval-augmented generation pipelines.
In the LLM eval & observability space, rag-inject-guard takes a focused approach. It focuses on detecting and preventing indirect prompt injection attacks hidden in RAG-retrieved documents before they reach the LLM. It is built as an open-source project for developers. rag-inject-guard is open source under the Apache-2.0 license. It runs on the command line.
Fevzi Ege Yurtsevenler builds and maintains rag-inject-guard, and it first shipped in 2026. The project is developed in the open on GitHub with 2 commits in the last 90 days. Key capabilities include Indirect Prompt Detection, Document Scanning, and Homoglyph Detection.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match