--- title: "agentdojo vs rebuff" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-protectai-rebuff" tools: ["ethz-spylab-agentdojo", "protectai-rebuff"] --- # agentdojo vs rebuff *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 rebuff if rebuff is a detector for prompt injection attacks on large language models and operates in TypeScript under an Apache-2.0 license. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [rebuff](https://playground.rebuff.ai) has 1.5k stars, 141 forks, and 33 open issues, last pushed Aug 7, 2024. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [rebuff's repository](https://github.com/protectai/rebuff). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [rebuff](/tools/protectai-rebuff.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | LLM Prompt Injection Detector | | Stars | 716 | 1,516 | | Forks | 188 | 141 | | Open issues | 41 | 33 | | Language | Python | TypeScript | | Adopt for | AgentDojo serves as a benchmarking environment to evaluate security attacks, like prompt injection, and defenses for Large Language Model (LLM) agents. | Rebuff is a detector for prompt injection attacks on large language models and operates in TypeScript under an Apache-2.0 license. | | Persona | - | - | | Runtime | - | - | | License | MIT | Apache-2.0 | | 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) | [rebuff](/tools/protectai-rebuff.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Archived (8%) | | Days since push | 63d | 727d | | Archived on GitHub | No | Yes | | Open issues (now) | 41 | 33 | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/protectai-rebuff/trust.md) | ## Shared compatibility - **Python**: [agentdojo](/tools/ethz-spylab-agentdojo.md) - Python runtime; [rebuff](/tools/protectai-rebuff.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: rebuff - **Adopt for:** Rebuff is a detector for prompt injection attacks on large language models and operates in TypeScript under an Apache-2.0 license. ## Choose when ### Choose agentdojo if… - agentdojo is primarily Python; rebuff is TypeScript. - License: agentdojo is MIT, rebuff is Apache-2.0. - 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. - 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 rebuff if… - rebuff is primarily TypeScript; agentdojo is Python. - License: rebuff is Apache-2.0, agentdojo is MIT. - Tags unique to rebuff: llm, llmops, prompt-engineering, prompts. - Use Rebuff when you need precise detection of prompt injection vulnerabilities specific to your deployment, especially if it relies heavily on interactions with 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 rebuff - Do not use Rebuff if setting up and managing multiple provider services like Supabase, OpenAI, Pinecone, or Chroma is inconvenient or infeasible for your project requirements. ## Common questions ### What is the difference between agentdojo and rebuff? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. rebuff: LLM Prompt Injection Detector. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over rebuff? Choose agentdojo over rebuff when agentdojo is primarily Python; rebuff is TypeScript; License: agentdojo is MIT, rebuff is Apache-2.0; 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; 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 rebuff over agentdojo? Choose rebuff over agentdojo when rebuff is primarily TypeScript; agentdojo is Python; License: rebuff is Apache-2.0, agentdojo is MIT; Tags unique to rebuff: llm, llmops, prompt-engineering, prompts; Use Rebuff when you need precise detection of prompt injection vulnerabilities specific to your deployment, especially if it relies heavily on interactions with 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 rebuff? Do not use Rebuff if setting up and managing multiple provider services like Supabase, OpenAI, Pinecone, or Chroma is inconvenient or infeasible for your project requirements. ### Is agentdojo or rebuff more popular on GitHub? rebuff has more GitHub stars (1,516 vs 716). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and rebuff open source? Yes - both are open-source projects on GitHub (agentdojo: MIT, rebuff: Apache-2.0). ### Where can I find alternatives to agentdojo or rebuff? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [rebuff alternatives](/tools/protectai-rebuff/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [rebuff markdown twin](/tools/protectai-rebuff/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-protectai-rebuff.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or rebuff? agentdojo: Steady. rebuff: Archived. 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 rebuff? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [rebuff trust report](/tools/protectai-rebuff/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/_