Intel

AIKIDO-2026-588943

lightrag-hku is vulnerable to Uncontrolled Resource Consumption

Uncontrolled Resource ConsumptionGHSA-r8jh-295g-vv42 Published Aug 11, 2026

75

High Risk

This Affects:

PYTHONlightrag-hku
0.0.1 - 1.5.5
Fixed in 1.5.6
Are you affected? Scan for Free

TL;DR

The LightRAG API server passes the chat message from the Ollama-compatible /api/chat route straight to a synchronous tokenizer call on the asyncio event loop with no size limit. The message field and message list are unbounded and the route is reachable without authentication under the shipped default whitelist, so a single large body ties up the only serving thread and can exhaust process memory. The fix caps message size, moves token counting off the event loop, and enforces a global request-body limit.

Who does this affect?

You are affected if you are using a version that falls within the vulnerable range and you expose the LightRAG API server.

Background info

lightrag-hku is vulnerable to Uncontrolled Resource Consumption in versions 0.0.1 - 1.5.5.

How to fix this

Upgrade the lightrag-hku library to the patch version.