{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/doomarena-a-framework-for-testing-ai-agents","title":"DoomArena: A framework for Testing AI Agents Against Evolving Security Threats","arxiv_id":"2504.14064","date":"2025-04-18","proceeding":null,"authors":["Leo Boisvert","Mihir Bansal","Chandra Kiran Reddy Evuru","Gabriel Huang","Abhay Puri","Avinandan Bose","Maryam Fazel","Quentin Cappart","Jason Stanley","Alexandre Lacoste","Alexandre Drouin","Krishnamurthy Dvijotham"],"abstract":"We present DoomArena, a security evaluation framework for AI agents. DoomArena is designed on three principles: 1) It is a plug-in framework and integrates easily into realistic agentic frameworks like BrowserGym (for web agents) and $\\tau$-bench (for tool calling agents); 2) It is configurable and allows for detailed threat modeling, allowing configuration of specific components of the agentic framework being attackable, and specifying targets for the attacker; and 3) It is modular and decouples the development of attacks from details of the environment in which the agent is deployed, allowing for the same attacks to be applied across multiple environments. We illustrate several advantages of our framework, including the ability to adapt to new threat models and environments easily, the ability to easily combine several previously published attacks to enable comprehensive and fine-grained security testing, and the ability to analyze trade-offs between various vulnerabilities and performance. We apply DoomArena to state-of-the-art (SOTA) web and tool-calling agents and find a number of surprising results: 1) SOTA agents have varying levels of vulnerability to different threat models (malicious user vs malicious environment), and there is no Pareto dominant agent across all threat models; 2) When multiple attacks are applied to an agent, they often combine constructively; 3) Guardrail model-based defenses seem to fail, while defenses based on powerful SOTA LLMs work better. DoomArena is available at https://github.com/ServiceNow/DoomArena.","url_abs":"https://arxiv.org/abs/2504.14064v2","url_pdf":"https://arxiv.org/pdf/2504.14064v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"links_only","authors_date_abstract":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license), from the Kaggle arXiv metadata snapshot of 2026-09-12"},"code_links":[{"paper_slug":"doomarena-a-framework-for-testing-ai-agents","repo_url":"https://github.com/servicenow/doomarena","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2504.14064","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.14064"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/servicenow/doomarena","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_violates":1,"unverified":5},"by_repo_kind":{"official":{"samples":6,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"7c508037b40522af","entry":"str2bool","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/core/src/doomarena/utils.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/core/src/doomarena/utils.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7c508037b40522af"}},{"code_sha256_prefix":"db029d935f1b22a2","entry":"get_banner_injection_script","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/browsergym/src/doomarena/browsergym/init_scripts.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/browsergym/src/doomarena/browsergym/init_scripts.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"db029d935f1b22a2"}},{"code_sha256_prefix":"7a0d2d77d6f59ad9","entry":"get_banner_update_script","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7a0d2d77d6f59ad9"}},{"code_sha256_prefix":"1ee09ed9e841cd10","entry":"get_popup_insertion_script","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1ee09ed9e841cd10"}},{"code_sha256_prefix":"2829a459708a8c31","entry":"get_results_path","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/mailinject/src/doomarena/mailinject/experiments.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/mailinject/src/doomarena/mailinject/experiments.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2829a459708a8c31"}},{"code_sha256_prefix":"334c9c4e58c08dbe","entry":"get_user_content_replacement_script","repo":"servicenow/doomarena","repo_kind":"official","path":"doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","file_url":"https://github.com/servicenow/doomarena/blob/HEAD/doomarena/browsergym/src/doomarena/browsergym/attack_scripts.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"334c9c4e58c08dbe"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}