ethz-spylab/agentdojo
Environment to evaluate prompt injection attacks on LLM agents
View on index · View in 3D Map
Related repos
- agentic_security: tools for scanning and securing LLMs against emerging threats in CI/CD workflows
- Discovers tools for autonomous agents
- LLM pentesting tool focused on prompt injection attacks
- Unified environment for developing and testing LLM agents with feedback
- Active tool discovery for autonomous LLM agents
- Scans Model Context Protocol servers for vulnerabilities
- Research papers on LLM agents, organized by taxonomy, covering methods, tools, security, applications, and survey insights
- Framework for evaluating LLM agents in Factorio
- cybersecurity multi-tool for pentesting, OSINT, network scanning, Discord, ethical hacking
- Benchmark evaluates LLMs as agents
