--- title: "agentdojo vs PentestGPT" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-greydgl-pentestgpt" tools: ["ethz-spylab-agentdojo", "greydgl-pentestgpt"] --- # agentdojo vs PentestGPT *GraphCanon updated Aug 17, 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 PentestGPT if pentestGPT specializes in automating parts of the penetration testing process through large language models, offering a unique approach to AI-assisted security assessments. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [PentestGPT](https://github.com/GreyDGL/PentestGPT) has 15k stars, 2.6k forks, and 65 open issues, last pushed Jul 14, 2026. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [PentestGPT's repository](https://github.com/GreyDGL/PentestGPT). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [PentestGPT](/tools/greydgl-pentestgpt.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | Automated Penetration Testing Agentic Framework Powered by Large Language Models | | Stars | 716 | 14,900 | | Forks | 188 | 2,604 | | Open issues | 41 | 65 | | 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. | PentestGPT specializes in automating parts of the penetration testing process through large language models, offering a unique approach to AI-assisted security assessments. | | Persona | - | - | | Runtime | - | - | | License | MIT | MIT License, which allows for free use, modification, and distribution provided that attribution is maintained and any warranties or liabilities are disclaimed. | | Categories | AI Agents, Evaluation & Observability | AI Agents, LLM Frameworks | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [PentestGPT](/tools/greydgl-pentestgpt.md) | | --- | --- | --- | | Days since push | 63d | 33d | | Open issues (now) | 41 | 65 | | Stars delta | Unknown | +597 (30d) | | Open issues delta | Unknown | +3 (30d) | | Owner type | Organization | User | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/greydgl-pentestgpt/trust.md) | ## 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: PentestGPT - **Requirements:** - Python environment; - Access to large language models as stipulated by PentestGPT's operational requirements - **Adopt for:** PentestGPT specializes in automating parts of the penetration testing process through large language models, offering a unique approach to AI-assisted security assessments. - **License detail:** MIT License, which allows for free use, modification, and distribution provided that attribution is maintained and any warranties or liabilities are disclaimed. ## 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, security. - Also covers Evaluation & Observability. - AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. ### Choose PentestGPT if… - Requirements: - Python environment; - Access to large language models as stipulated by PentestGPT's operational requirements. - Tags unique to PentestGPT: llm, penetration-testing, python. - Also covers LLM Frameworks. - PentestGPT ships Docker support for self-hosted deployment. - - When you need an automated framework for certain tasks within penetration testing that can be handled by large language models. ## 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 PentestGPT - - Avoid using PentestGPT if manual, nuanced analysis is required, as its reliance on LLM might not cover all complexities of a security assessment. - - If your organization does not have the legal authority to conduct penetration testing on specific targets, as indicated by its disclaimer for 'educational purposes and authorized security testing'. ## Common questions ### What is the difference between agentdojo and PentestGPT? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. PentestGPT: Automated Penetration Testing Agentic Framework Powered by Large Language Models. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over PentestGPT? Choose agentdojo over PentestGPT 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, security; Also covers Evaluation & Observability; 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 PentestGPT over agentdojo? Choose PentestGPT over agentdojo when Requirements: - Python environment; - Access to large language models as stipulated by PentestGPT's operational requirements; Tags unique to PentestGPT: llm, penetration-testing, python; Also covers LLM Frameworks; PentestGPT ships Docker support for self-hosted deployment; - When you need an automated framework for certain tasks within penetration testing that can be handled by large language models. ### 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 PentestGPT? - Avoid using PentestGPT if manual, nuanced analysis is required, as its reliance on LLM might not cover all complexities of a security assessment. - If your organization does not have the legal authority to conduct penetration testing on specific targets, as indicated by its disclaimer for 'educational purposes and authorized security testing'. ### Is agentdojo or PentestGPT more popular on GitHub? PentestGPT has more GitHub stars (14,900 vs 716). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and PentestGPT open source? Yes - both are open-source projects on GitHub (agentdojo: MIT, PentestGPT: MIT). ### Where can I find alternatives to agentdojo or PentestGPT? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [PentestGPT alternatives](/tools/greydgl-pentestgpt/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [PentestGPT markdown twin](/tools/greydgl-pentestgpt/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-greydgl-pentestgpt.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or PentestGPT? agentdojo: Steady. PentestGPT: Steady. 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 PentestGPT? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [PentestGPT trust report](/tools/greydgl-pentestgpt/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/_