--- title: "agentdojo vs AutoDefense" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-xhmy-autodefense" tools: ["ethz-spylab-agentdojo", "xhmy-autodefense"] --- # agentdojo vs AutoDefense *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 AutoDefense if autoDefense uses a multi-agent framework to mitigate jailbreak attacks on LLMs, installed via Python. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [AutoDefense](https://arxiv.org/abs/2403.04783) has 68 stars, 20 forks, and 1 open issues, last pushed Jan 15, 2026. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [AutoDefense's repository](https://github.com/XHMY/AutoDefense). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [AutoDefense](/tools/xhmy-autodefense.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | Multi-Agent LLM Defense against Jailbreak Attacks | | Stars | 716 | 68 | | Forks | 188 | 20 | | Open issues | 41 | 1 | | 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. | AutoDefense uses a multi-agent framework to mitigate jailbreak attacks on LLMs, installed via Python. | | Persona | - | - | | Runtime | - | - | | License | MIT | MIT | | Categories | AI Agents, Evaluation & Observability | AI Agents, Evaluation & Observability | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [AutoDefense](/tools/xhmy-autodefense.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Slowing (36%) | | Days since push | 63d | 201d | | Open issues (now) | 41 | 1 | | Owner type | Organization | User | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/xhmy-autodefense/trust.md) | ## Shared compatibility - **Python**: [agentdojo](/tools/ethz-spylab-agentdojo.md) - Python runtime; [AutoDefense](/tools/xhmy-autodefense.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: AutoDefense - **Adopt for:** AutoDefense uses a multi-agent framework to mitigate jailbreak attacks on LLMs, installed via Python. ## 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, prompt-injection. - AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### Choose AutoDefense if… - Tags unique to AutoDefense: defense-mechanism, jailbreak prevention, llm-defense, multi-agent. - Implementing robust defenses for enterprise-level AI projects with high-security requirements - Leaner open-issue backlog (1). ## 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 AutoDefense - Projects requiring light-weight solutions where multi-agent systems might introduce complexity overhead - Environments without access to Python and its ecosystem, as AutoDefense depends on specific Python packages ## Common questions ### What is the difference between agentdojo and AutoDefense? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over AutoDefense? Choose agentdojo over AutoDefense 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, prompt-injection; 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 AutoDefense over agentdojo? Choose AutoDefense over agentdojo when Tags unique to AutoDefense: defense-mechanism, jailbreak prevention, llm-defense, multi-agent; Implementing robust defenses for enterprise-level AI projects with high-security requirements; Leaner open-issue backlog (1). ### 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 AutoDefense? Projects requiring light-weight solutions where multi-agent systems might introduce complexity overhead Environments without access to Python and its ecosystem, as AutoDefense depends on specific Python packages ### Is agentdojo or AutoDefense more popular on GitHub? agentdojo has more GitHub stars (716 vs 68). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and AutoDefense open source? Yes - both are open-source projects on GitHub (agentdojo: MIT, AutoDefense: MIT). ### Where can I find alternatives to agentdojo or AutoDefense? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [AutoDefense alternatives](/tools/xhmy-autodefense/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [AutoDefense markdown twin](/tools/xhmy-autodefense/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-xhmy-autodefense.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or AutoDefense? agentdojo: Steady. AutoDefense: Slowing. 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 AutoDefense? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [AutoDefense trust report](/tools/xhmy-autodefense/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/_