155 agent-dojo(1) - Kailash Tool Reference
155.1 agent-dojo
agent-dojo - AgentDojo — ETH’s dynamic evaluation environment for prompt-injection attacks and defenses on LLM agents (upstream repo agentdojo).
| Layer | ATTACK |
| Category | a-03-agentic-mcp-tool-use |
| Safety class | safe |
| Licence | MIT |
| Upstream | github.com/ethz-spylab/agentdojo |
155.1.1 NAME
agent-dojo - AgentDojo — ETH’s dynamic evaluation environment for prompt-injection attacks and defenses on LLM agents (upstream repo agentdojo).
155.1.2 SYNOPSIS & usage
Derived from the upstream README at generation time; verify against --help where the tool is installed.
pip install agentdojopip install "agentdojo[transformers]"155.1.3 DESCRIPTION (upstream)
Edoardo Debenedetti1, Jie Zhang1, Mislav Balunović1,2, Luca Beurer-Kellner1,2, Marc Fischer1,2, Florian Tramèr1 pip install agentdojo > [!IMPORTANT] > Note that the API of the package is still under development and might change in the future. If you want to use the prompt injection detector, you need to install the transformers extra: pip install “agentdojo[transformers]” The benchmark can be run with the [benchmark](src/agen
155.1.4 SEE ALSO
- Category: a-03-agentic-mcp-tool-use in the category directory (sister tools listed there).
ref: v0.1.35-unstable-2026-06-02 - verified 2026-10-14