# AgentDojo

AgentDojo is an AI benchmark; this page records what it measures and traces its data, grader, and scores to primary sources.

Evaluates agent task utility and resistance to prompt injection while agents use tools over untrusted data.

## Details

- Address: Online, Global
- Website: [agentdojo.spylab.ai/](<https://agentdojo.spylab.ai/>)
- Tags: Helpful, Viral, Science
- Category: Safety
- Status: Active
- Data licence: MIT
- Data access: open
- Grader: simulation
- Organization: ETH Zurich
- Released: 2024-02
- Paper: [arxiv.org/abs/2406.13352](<https://arxiv.org/abs/2406.13352>)
- Leaderboard: [agentdojo.spylab.ai/results/](<https://agentdojo.spylab.ai/results/>)
- Repository: [github.com/ethz-spylab/agentdojo](<https://github.com/ethz-spylab/agentdojo>)
- Task count: 97
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is AgentDojo?

AgentDojo is at Online, Global.

### How do I contact AgentDojo?

To contact AgentDojo, visit https://agentdojo.spylab.ai/.

## Source

From [Benchmark primary sources](https://aibenchmarkindex.com/about), license CC0-1.0, checked 2026-10-15.

Last updated October 15, 2026.

## Links

- This benchmark on the [web page](<https://aibenchmarkindex.com/global/online/agentdojo>)
- [Safety in Online](<https://aibenchmarkindex.com/global/online/category/safety.md>)
- [More benchmarks in Online](<https://aibenchmarkindex.com/global/online.md>)

> **Check before you compare.** Benchmark methods and leaderboards change. Follow the cited primary source before using a result in a decision.