AgentLeak (IEEE Access 2026): benchmark for internal-channel privacy leakage in multi-agent LLM systems. Inter-agent messages leak 68.8% vs 27.2% in final outputs; output-only audits miss 41.7%. 1,000 scenarios, 7 channels, 5 LLMs. Tool & MCP server: github.com/yagobski/agentleak
☆45Oct 5, 2026Updated last month
Alternatives and similar repositories for AgentLeak
Users that are interested in AgentLeak are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- The repository of the NeurIPS25 paper: Contextual Integrity in LLMs via Reasoning and Reinforcement Learning.☆16Jun 8, 2026Updated 5 months ago
- This is the code repo for the paper AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play (NeurIPS 2025 Spotl…☆25Sep 29, 2025Updated last year
- github for Beyond ReAct: A Planner-Centric Framework for Complex \\ Tool-Augmented LLM Reasoning☆19Feb 27, 2026Updated 8 months ago
- SocioBench: Modeling Human Behavior in Sociological Surveys with Large Language Models☆13Aug 27, 2026Updated 2 months ago
- LobotoMl is a set of scripts and tools to assess production deployments of ML services☆10May 16, 2022Updated 4 years ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- Towards Automating Data Access Permissions in AI Agents☆15Jan 27, 2026Updated 9 months ago
- Official Implementation of "ToolSafe: Enhancing Tool Invocation Safety of LLM-based Agents via Proactive Step-level Guardrail and Feedbac…☆87Nov 5, 2026Updated 2 weeks ago
- Bundle of security analysis scripts for keras tensorflow models☆16Apr 15, 2024Updated 2 years ago
- Generative web directory fuzzer,crawling and subdomain checker based on chatgpt☆15May 15, 2024Updated 2 years ago
- A collection of sample code used in some experiments with Sliver C2☆17Mar 28, 2023Updated 3 years ago
- [ACL 2025] The official code for "AGrail: A Lifelong Agent Guardrail with Effective and Adaptive Safety Detection".☆47Aug 12, 2026Updated 3 months ago
- Official code for the paper: "Multi-User Large Language Model Agents"☆38Nov 9, 2026Updated last week
- Harness-Delta Attribution: Dissecting Gains From Harness Evolution☆16Oct 19, 2026Updated last month
- A security-first linter for code that shouldn't need linting☆19Sep 12, 2023Updated 3 years ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- CVE-2023-50254: PoC Exploit for Deepin-reader RCE that affects unpatched Deepin Linux Desktops. Deepin Linux's default document reader "d…☆16Dec 22, 2023Updated 2 years ago
- The official training code of SingProbe.☆35Sep 3, 2026Updated 2 months ago
- [ACL 2026] Hard to Read, Easy to Jailbreak: How Visual Degradation Bypasses MLLM Safety Alignment☆14May 11, 2026Updated 6 months ago
- Kelp is a novel plug-in framework that enables streaming risk detection within the LM generation pipeline.☆12Oct 20, 2025Updated last year
- exploiting and defending neural networks(神经网络攻防专栏)☆15Mar 2, 2021Updated 5 years ago
- FedE4RAG☆21Dec 15, 2025Updated 11 months ago
- ☆57Mar 24, 2026Updated 7 months ago
- The official implementation of the paper "AgentLAB: Benchmarking LLM Agents against Long-Horizon Attacks"☆55Jun 1, 2026Updated 5 months ago
- Code for "Neural Retrievers are Biased Towards LLM-Generated Content"☆14Oct 18, 2024Updated 2 years ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- Safety and privacy with LLMs☆14Sep 25, 2023Updated 3 years ago
- Can Agents Write Secure and Correct Code?☆21Updated this week
- TraitBasis applied to TauBench☆19Oct 22, 2026Updated 3 weeks ago
- Effective Unsupervised Domain Adaptation of Neural Rankers by Diversifying Synthetic Query Generation☆16Apr 23, 2025Updated last year
- Official code repository for "A2ASecBench: A Protocol-Aware Security Benchmark for Agent-to-Agent Multi-Agent Systems" at ICLR 2026.☆12May 8, 2026Updated 6 months ago
- ☆51May 29, 2026Updated 5 months ago
- [CVPR 2023] "TrojViT: Trojan Insertion in Vision Transformers" by Mengxin Zheng, Qian Lou, Lei Jiang☆15Jan 5, 2024Updated 2 years ago
- mcp server for yourware☆10Mar 23, 2026Updated 7 months ago
- We introduce Multi-P2A, a Multi-perspective Privacy Assessment aimed at evaluating privacy risks of LVLMs comprehensively.☆13Sep 23, 2026Updated last month
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- [ICLR 2026] Fewer Weights, More Problems: A Practical Attack on LLM Pruning☆14Apr 6, 2026Updated 7 months ago
- 📄 [Talk] OFFZONE 2022 / ODS Data Halloween 2022: Black-box attacks on ML models + with use of open-source tools☆14May 23, 2023Updated 3 years ago
- Official PyTorch implementation of "MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks"☆20Dec 4, 2025Updated 11 months ago
- Revisiting Cache Side-Channel Attacks in Deep Neural Networks Executables☆14Aug 27, 2024Updated 2 years ago
- ☆16Jun 15, 2024Updated 2 years ago
- AI Security Research☆17Jun 21, 2023Updated 3 years ago
- DynAuditClaw — A security audit skill that dynamically discovers your OpenClaw agent's real configuration, designs targeted attack scenar…☆15Apr 6, 2026Updated 7 months ago
Are these results useful?