FEAT: Add 0DIN threat feed dataset loader (0din_threatfeed) (#2034)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: romanlutz

doc/bibliography.md

@@ -5,6 +5,6 @@ All academic papers, research blogs, and technical reports referenced throughout
 :::{dropdown} Citation Keys
 :class: hidden-citations
 
-[@aakanksha2024multilingual; @adversaai2023universal; @andriushchenko2024tense; @anthropic2024manyshot; @aqrawi2024singleturncrescendo; @atr2026; @bethany2024mathprompt; @bhardwaj2023harmfulqa; @bhardwaj2024homer; @brahman2024coconot; @bryan2025agentictaxonomy; @bullwinkel2025airtlessons; @bullwinkel2025repeng; @bullwinkel2026trigger; @chao2023pair; @chao2024jailbreakbench; @cui2024orbench; @darkbench2025; @derczynski2024garak; @ding2023wolf; @embracethered2024unicode; @embracethered2025sneakybits; @gehman2020realtoxicityprompts; @ghosh2025aegis; @ghosh2025ailuminate; @gong2025figstep; @gupta2024walledeval; @haider2024phi3safety; @han2024medsafetybench; @hines2024spotlighting; @ji2023beavertails; @ji2024pkusaferlhf; @jiang2025sosbench; @jones2025computeruse; @kingma2014adam; @li2024mossbench; @li2024saladbench; @li2024wmdp; @lin2023toxicchat; @liu2024flipattack; @liu2024mmsafetybench; @lopez2024pyrit; @luo2024jailbreakv; @lv2024codechameleon; @mazeika2023tdc; @mazeika2024harmbench; @mckee2024transparency; @mehrotra2023tap; @microsoft2024skeletonkey; @palaskar2025vlsu; @pfohl2024equitymedqa; @promptfoo2025ccp; @robustintelligence2024bypass; @roccia2024promptintel; @rottger2023xstest; @rottger2025msts; @russinovich2024crescendo; @russinovich2025price; @scheuerman2025transphobia; @shaikh2022second; @shayegani2025computeruse; @shen2023donotanything; @sheshadri2024lat; @souly2024strongreject; @stok2023ansi; @tan2026comicjailbreak; @tang2025multilingual; @tedeschi2024alert; @vantaylor2024socialbias; @vidgen2023simplesafetytests; @wang2023decodingtrust; @wang2023donotanswer; @wang2025siuo; @wei2023jailbroken; @xie2024sorrybench; @yu2023gptfuzzer; @yuan2023cipherchat; @zeng2024persuasion; @zhang2024cbtbench; @ziems2022mic; @zou2023gcg]
+[@aakanksha2024multilingual; @adversaai2023universal; @andriushchenko2024tense; @anthropic2024manyshot; @aqrawi2024singleturncrescendo; @atr2026; @bethany2024mathprompt; @bhardwaj2023harmfulqa; @bhardwaj2024homer; @brahman2024coconot; @bryan2025agentictaxonomy; @bullwinkel2025airtlessons; @bullwinkel2025repeng; @bullwinkel2026trigger; @chao2023pair; @chao2024jailbreakbench; @cui2024orbench; @darkbench2025; @derczynski2024garak; @ding2023wolf; @embracethered2024unicode; @embracethered2025sneakybits; @gehman2020realtoxicityprompts; @ghosh2025aegis; @ghosh2025ailuminate; @gong2025figstep; @gupta2024walledeval; @haider2024phi3safety; @han2024medsafetybench; @hines2024spotlighting; @inie2025summon; @ji2023beavertails; @ji2024pkusaferlhf; @jiang2025sosbench; @jones2025computeruse; @kingma2014adam; @li2024mossbench; @li2024saladbench; @li2024wmdp; @lin2023toxicchat; @liu2024flipattack; @liu2024mmsafetybench; @lopez2024pyrit; @luo2024jailbreakv; @lv2024codechameleon; @mazeika2023tdc; @mazeika2024harmbench; @mckee2024transparency; @mehrotra2023tap; @microsoft2024skeletonkey; @odin2024; @palaskar2025vlsu; @pfohl2024equitymedqa; @promptfoo2025ccp; @robustintelligence2024bypass; @roccia2024promptintel; @rottger2023xstest; @rottger2025msts; @russinovich2024crescendo; @russinovich2025price; @scheuerman2025transphobia; @shaikh2022second; @shayegani2025computeruse; @shen2023donotanything; @sheshadri2024lat; @souly2024strongreject; @stok2023ansi; @tan2026comicjailbreak; @tang2025multilingual; @tedeschi2024alert; @vantaylor2024socialbias; @vidgen2023simplesafetytests; @wang2023decodingtrust; @wang2023donotanswer; @wang2025siuo; @wei2023jailbroken; @xie2024sorrybench; @yu2023gptfuzzer; @yuan2023cipherchat; @zeng2024persuasion; @zhang2024cbtbench; @ziems2022mic; @zou2023gcg]
 
 :::

doc/code/datasets/1_loading_datasets.ipynb

@@ -14,6 +14,7 @@
     "The following command lists all built-in datasets available in PyRIT. Some datasets are stored locally, while others are fetched remotely from sources like HuggingFace.\n",
     "\n",
     "Many of these datasets come from published research, including\n",
+    "0DIN [@odin2024],\n",
     "Aegis [@ghosh2025aegis],\n",
     "Agent Threat Rules [@atr2026],\n",
     "ALERT [@tedeschi2024alert],\n",
@@ -74,6 +75,7 @@
        " '0din_incremental_table_completion',\n",
        " '0din_placeholder_injection',\n",
        " '0din_technical_field_guide',\n",
+       " '0din_threatfeed',\n",
        " 'adv_bench',\n",
        " 'aegis_content_safety',\n",
        " 'agent_threat_rules',\n",

doc/code/datasets/1_loading_datasets.py

@@ -18,6 +18,7 @@
 # The following command lists all built-in datasets available in PyRIT. Some datasets are stored locally, while others are fetched remotely from sources like HuggingFace.
 #
 # Many of these datasets come from published research, including
+# 0DIN [@odin2024],
 # Aegis [@ghosh2025aegis],
 # Agent Threat Rules [@atr2026],
 # ALERT [@tedeschi2024alert],

doc/references.bib

@@ -88,6 +88,22 @@ @misc{roccia2024promptintel
   url       = {https://promptintel.novahunting.ai/feed},
 }
 
+@misc{odin2024,
+  title     = {{0DIN}: {GenAI} Bug Bounty and Threat Feed},
+  author    = {{Mozilla 0DIN}},
+  year      = {2024},
+  url       = {https://0din.ai/},
+  note      = {0DIN Jailbreak / Threat Feed},
+}
+
+@article{inie2025summon,
+  title     = {Summon a Demon and Bind it: A Grounded Theory of {LLM} Red Teaming},
+  author    = {Nanna Inie and Jonathan Stray and Leon Derczynski},
+  journal   = {PLoS ONE},
+  year      = {2025},
+  url       = {https://arxiv.org/abs/2311.06237},
+}
+
 @misc{vantaylor2024socialbias,
   title     = {A Red-Teaming Repository of Existing Social Bias Prompts},
   author    = {Simone Van Taylor},

pyrit/datasets/seed_datasets/remote/__init__.py

@@ -119,6 +119,12 @@
 from pyrit.datasets.seed_datasets.remote.multilingual_vulnerability_dataset import (
     _MultilingualVulnerabilityDataset,
 )
+from pyrit.datasets.seed_datasets.remote.odin_dataset import (
+    ODINSecurityBoundary,
+    ODINSeverity,
+    ODINTaxonomyCategory,
+    _ODINDataset,
+)
 from pyrit.datasets.seed_datasets.remote.or_bench_dataset import (
     _ORBench80KDataset,
     _ORBenchHardDataset,
@@ -199,6 +205,9 @@
     "MMSafetyBenchCategory",
     "MMSafetyBenchVariant",
     "MossBenchOversensitivityType",
+    "ODINSecurityBoundary",
+    "ODINSeverity",
+    "ODINTaxonomyCategory",
     "PromptIntelCategory",
     "PromptIntelSeverity",
     "SGXSTestLabel",
@@ -244,6 +253,7 @@
     "_MossBenchDataset",
     "_MSTSDataset",
     "_MultilingualVulnerabilityDataset",
+    "_ODINDataset",
     "_ORBench80KDataset",
     "_ORBenchHardDataset",
     "_ORBenchToxicDataset",