ethz-spylab/agentdojo
A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.
Provides modular benchmarking for prompt injection attacks across multiple task suites with configurable defenses (tool filtering, input sanitization) and attack strategies (tool knowledge, context awareness). Built as a Python package with pluggable LLM backends and a results registry for standardized evaluation and comparison of agent robustness against adversarial inputs.
471 stars and 15,222 monthly downloads. Available on PyPI.
Stars
471
Forks
118
Language
Python
License
MIT
Category
Last pushed
Mar 12, 2026
Monthly downloads
15,222
Commits (30d)
0
Dependencies
14
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/ethz-spylab/agentdojo"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
guardrails-ai/guardrails
Adding guardrails to large language models.
JasonLovesDoggo/caddy-defender
Caddy module to block or manipulate requests originating from AIs or cloud services trying to...
inkdust2021/VibeGuard
Uses just 1% memory while protecting 99% of your personal privacy.
Heiberg-Industries/designbrief
Design guardrails, not templates. A library of UI design direction files that give LLMs (and...
ankitlade12/AgentArmor
The full-stack safety layer for AI agents. Budget limits, prompt injection shields, PII...