--- title: "agentdojo vs jailbreakbench" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-jailbreakbench-jailbreakbench" tools: ["ethz-spylab-agentdojo", "jailbreakbench-jailbreakbench"] --- # agentdojo vs jailbreakbench *GraphCanon updated Aug 5, 2026* ## Verdict Pick agentdojo if agentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents; pick jailbreakbench if jailbreakBench is an open robustness benchmark specifically designed to evaluate language models against jailbreaking attacks. It aims to quantify the resilience of language models under adversarial conditions. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [jailbreakbench](https://jailbreakbench.github.io) has 645 stars, 75 forks, and 11 open issues, last pushed Apr 4, 2025. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [jailbreakbench's repository](https://github.com/JailbreakBench/jailbreakbench). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [jailbreakbench](/tools/jailbreakbench-jailbreakbench.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | An Open Robustness Benchmark for Jailbreaking Language Models | | Stars | 716 | 645 | | Forks | 188 | 75 | | Open issues | 41 | 11 | | Language | Python | Python | | Adopt for | AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. | JailbreakBench is an open robustness benchmark specifically designed to evaluate language models against jailbreaking attacks. It aims to quantify the resilience of language models under adversarial conditions. | | Persona | - | - | | Runtime | - | - | | License | MIT | MIT | | Categories | AI Agents, Evaluation & Observability | Evaluation & Observability | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [jailbreakbench](/tools/jailbreakbench-jailbreakbench.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Dormant (18%) | | Days since push | 63d | 487d | | Open issues (now) | 41 | 11 | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/jailbreakbench-jailbreakbench/trust.md) | ## Shared compatibility - **Python**: [agentdojo](/tools/ethz-spylab-agentdojo.md) - Python runtime; [jailbreakbench](/tools/jailbreakbench-jailbreakbench.md) - Python runtime ## Decision facts: agentdojo - **Pricing:** freemium - Open-source under the MIT License. Some advanced features might require additional libraries or APIs. - **Requirements:** Min 8 GB RAM - **Adopt for:** AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ## Decision facts: jailbreakbench - **Adopt for:** JailbreakBench is an open robustness benchmark specifically designed to evaluate language models against jailbreaking attacks. It aims to quantify the resilience of language models under adversarial conditions. ## Choose when ### Choose agentdojo if… - Pricing: Open-source under the MIT License. Some advanced features might require additional libraries or APIs.. - Requirements: Min 8 GB RAM. - Tags unique to agentdojo: benchmark, large language models, prompt-injection, security. - Also covers AI Agents. - AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### Choose jailbreakbench if… - Tags unique to jailbreakbench: jailbreaking, language-models, neurips-2024-datasets-and-benchmarks-tra, robustness-benchmark. - JailbreakBench is an open robustness benchmark specifically designed to evaluate language models against jailbreaking attacks. It aims to quantify the resilience of language models under adversarial conditions. - Leaner open-issue backlog (11). ## When NOT to use agentdojo - AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. - Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ## When NOT to use jailbreakbench - Last GitHub push was 508 days ago (dormant maintenance, Apr 4, 2025). Validate activity before betting a new project on jailbreakbench. - Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ## Common questions ### What is the difference between agentdojo and jailbreakbench? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. jailbreakbench: An Open Robustness Benchmark for Jailbreaking Language Models. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over jailbreakbench? Choose agentdojo over jailbreakbench when Pricing: Open-source under the MIT License. Some advanced features might require additional libraries or APIs.; Requirements: Min 8 GB RAM; Tags unique to agentdojo: benchmark, large language models, prompt-injection, security; Also covers AI Agents; AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### When should I choose jailbreakbench over agentdojo? Choose jailbreakbench over agentdojo when Tags unique to jailbreakbench: jailbreaking, language-models, neurips-2024-datasets-and-benchmarks-tra, robustness-benchmark; JailbreakBench is an open robustness benchmark specifically designed to evaluate language models against jailbreaking attacks. It aims to quantify the resilience of language models under adversarial conditions; Leaner open-issue backlog (11). ### When should I avoid agentdojo? AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ### When should I avoid jailbreakbench? Last GitHub push was 508 days ago (dormant maintenance, Apr 4, 2025). Validate activity before betting a new project on jailbreakbench. Evaluation & Observability: Defer heavyweight eval infra only until you have real traffic - never skip it once users depend on answers. ### Is agentdojo or jailbreakbench more popular on GitHub? agentdojo has more GitHub stars (716 vs 645). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and jailbreakbench open source? Yes - both are open-source projects on GitHub (agentdojo: MIT, jailbreakbench: MIT). ### Where can I find alternatives to agentdojo or jailbreakbench? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [jailbreakbench alternatives](/tools/jailbreakbench-jailbreakbench/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [jailbreakbench markdown twin](/tools/jailbreakbench-jailbreakbench/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/ethz-spylab-agentdojo-vs-jailbreakbench-jailbreakbench.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or jailbreakbench? agentdojo: Steady. jailbreakbench: Dormant. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for agentdojo and jailbreakbench? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [jailbreakbench trust report](/tools/jailbreakbench-jailbreakbench/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=ethz-spylab-agentdojo`](/api/graphcanon/graph?tool=ethz-spylab-agentdojo) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_