AI-secure/AgentPoison

[NeurIPS 2024] Official implementation for "AgentPoison: Red-teaming LLM Agents via Memory or Knowledge Base Backdoor Poisoning"

45
/ 100
Emerging

Introduces gradient-guided trigger optimization targeting RAG embedders (BERT, DPR, ANCE, BGE, REALM, ORQA) with coherence filtering and configurable poisoning strategies across multiple agent architectures (autonomous driving, QA, EHR systems). Demonstrates backdoor attacks on agent memory and knowledge retrieval by crafting adversarial passage tokens that manipulate embedding similarity scores while maintaining semantic coherence through perplexity filtering.

203 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

203

Forks

27

Language

Python

License

MIT

Last pushed

Apr 12, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/AI-secure/AgentPoison"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.