--- title: "agentdojo vs DevEval" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/ethz-spylab-agentdojo-vs-open-compass-deveval" tools: ["ethz-spylab-agentdojo", "open-compass-deveval"] --- # agentdojo vs DevEval *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 DevEval if devEval suits organizations requiring Python-centric software development benchmarks and practices evaluation. [agentdojo](https://agentdojo.spylab.ai/) reports 716 GitHub stars, 188 forks, and 41 open issues, last pushed Jun 2, 2026. [DevEval](https://github.com/open-compass/DevEval) has 138 stars, 13 forks, and 0 open issues, last pushed May 30, 2024. Figures are from public GitHub metadata via [agentdojo's repository](https://github.com/ethz-spylab/agentdojo) and [DevEval's repository](https://github.com/open-compass/DevEval). | | [agentdojo](/tools/ethz-spylab-agentdojo.md) | [DevEval](/tools/open-compass-deveval.md) | | --- | --- | --- | | Tagline | A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents | A Comprehensive Benchmark for Software Development | | Stars | 716 | 138 | | Forks | 188 | 13 | | Open issues | 41 | 0 | | 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. | DevEval suits organizations requiring Python-centric software development benchmarks and practices evaluation. | | 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) | [DevEval](/tools/open-compass-deveval.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Dormant (18%) | | Days since push | 63d | 797d | | Open issues (now) | 41 | 0 | | Full report | [trust report](/tools/ethz-spylab-agentdojo/trust.md) | [trust report](/tools/open-compass-deveval/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: DevEval - **Adopt for:** DevEval suits organizations requiring Python-centric software development benchmarks and practices evaluation. ## Choose when ### Choose agentdojo if… - License: agentdojo is MIT, DevEval 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: 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 DevEval if… - License: DevEval is Apache-2.0, agentdojo is MIT. - Tags unique to DevEval: docker-supported, python, software-development. - Choose DevEval when you require comprehensive benchmarking specifically for software development using Python. ## 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 DevEval - Avoid DevEval if your software projects heavily rely on languages other than Python. - Do not use it when a non-Dockerized evaluation tool is preferred due to organizational constraints or preferences. ## Common questions ### What is the difference between agentdojo and DevEval? agentdojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents. DevEval: A Comprehensive Benchmark for Software Development. See the comparison table for live GitHub stats and shared categories. ### When should I choose agentdojo over DevEval? Choose agentdojo over DevEval when License: agentdojo is MIT, DevEval 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: 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 DevEval over agentdojo? Choose DevEval over agentdojo when License: DevEval is Apache-2.0, agentdojo is MIT; Tags unique to DevEval: docker-supported, python, software-development; Choose DevEval when you require comprehensive benchmarking specifically for software development using Python. ### 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 DevEval? Avoid DevEval if your software projects heavily rely on languages other than Python. Do not use it when a non-Dockerized evaluation tool is preferred due to organizational constraints or preferences. ### Is agentdojo or DevEval more popular on GitHub? agentdojo has more GitHub stars (716 vs 138). Stars measure visibility, not whether either tool fits your constraints. ### Are agentdojo and DevEval open source? Yes - both are open-source projects on GitHub (agentdojo: MIT, DevEval: Apache-2.0). ### Where can I find alternatives to agentdojo or DevEval? GraphCanon lists graph-backed alternatives at [agentdojo alternatives](/tools/ethz-spylab-agentdojo/alternatives) and [DevEval alternatives](/tools/open-compass-deveval/alternatives) ([agentdojo markdown twin](/tools/ethz-spylab-agentdojo/alternatives.md), [DevEval markdown twin](/tools/open-compass-deveval/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-open-compass-deveval.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, agentdojo or DevEval? agentdojo: Steady. DevEval: 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 DevEval? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [agentdojo trust report](/tools/ethz-spylab-agentdojo/trust); [DevEval trust report](/tools/open-compass-deveval/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/_