Stealthy Prompt Injection and Poisoning in RAG Systems via Vector Database Embeddings
-
Updated
Nov 24, 2025 - Python
#
prompt-poisoning
Here are 3 public repositories matching this topic...
A CDN side advanced LLM poisoning technique has been applied to this otherwise innocuous project generated by Spark. Assist us in identifying the precise means by which the reasoning chain had been intercepted and/or evidence of token or string injection into the LLM's final outputs.
poisoning poisoning-attacks llm-inference token-poisoning cdn-side-exploit prompt-poisoning chain-of-reasoning-interception
-
Updated
Nov 16, 2025 - TypeScript
A CDN side advanced LLM poisoning technique was applied to this otherwise innocuous project generated by Spark. Assist us in identifying the precise means by which the reasoning chain had been intercepted and/or evidence of token or string injection into the LLM's final outputs.
poisoning-attacks llm-inference token-poisoning cdn-side-exploit prompt-poisoning chain-of-reasoning-interception llm-inference-poisoning
-
Updated
Nov 16, 2025 - TypeScript
Improve this page
Add a description, image, and links to the prompt-poisoning topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the prompt-poisoning topic, visit your repo's landing page and select "manage topics."