agentdojo
A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
GraphCanon updated Oct 6, 2026 · GitHub synced Oct 6, 2026
119views this month
Decision brief
AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents.
- Pricing:
- freemium - Open-source under the MIT License. Some advanced features might require additional libraries or APIs.
- Requirements:
- Min 8 GB RAM
Observed Jul 12, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Slowing (125d since push)
- As of Oct 6, 2026
- Provenance
- Not a fork · Organization account
- As of Oct 6, 2026
- Security (OSV)
- No lockfile
- As of Jul 11, 2026
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install agentdojoSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
AgentDojo provides a benchmarking environment to evaluate security attacks (like prompt injection) and defenses for large language model agents.
Capability facts
- Languages
- python
Source: github.language+pyproject.toml · Oct 6, 2026
Categories
Compatibility
Sourced claims from the README excerpt - not unsourced marketing copy.
Source: README excerpt (regex_v1, Oct 6, 2026)
python -m agentdojo.scripts.benchmark -s workspace -ut user_task_0 \Source link
Tags
README
<center AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents Edoardo Debenedetti<sup 1</sup , Jie Zhang<sup 1</sup , Mislav Balunović<sup 1,2</sup , Luca Beurer Kellner<sup 1,2</sup , Marc Fischer<sup 1,2</sup , Florian Tramèr<sup 1</s...
For agents
This page has a .md twin and JSON over the API.