1 line
27 KiB
JSON
1 line
27 KiB
JSON
{"repo": "andremmfaria/rexis", "files": {"config/.secrets_template.toml": {"95c6933ae08544d6b66a2e9567ff7ea1e68a9ec7": {"t": 1785745356.6738632, "n": 359, "h": "986fddb02604605b", "c": "db_password = \"super_secret_password\"\n\n# API keys used by Dynaconf via config/settings.toml (@get ...)\nopenai_api_key = \"sk-...\"\ndeepseek_api_key = \"dseek-...\"\n\n# NOTE: key name must be exactly 'malware_bazaar_api_key'\nmalware_bazaar_api_key = \"malw-bazaar-...\"\nvirus_total_api_key = \"vt-...\"\n\n# Optional/unused placeholders\ndecompiler_api_key = \"decomp-...\"\n"}, "70e2cc0aef1a5199cac798e7a103376b57d64907": {"t": 1785750639.5051239, "n": 359, "h": "986fddb02604605b", "c": "db_password = \"super_secret_password\"\n\n# API keys used by Dynaconf via config/settings.toml (@get ...)\nopenai_api_key = \"sk-...\"\ndeepseek_api_key = \"dseek-...\"\n\n# NOTE: key name must be exactly 'malware_bazaar_api_key'\nmalware_bazaar_api_key = \"malw-bazaar-...\"\nvirus_total_api_key = \"vt-...\"\n\n# Optional/unused placeholders\ndecompiler_api_key = \"decomp-...\"\n"}, "dfad596a6f712ba6be6bb70825251d826c0fd163": {"t": 1785750641.0845685, "n": 213, "h": "ac2c42394f4447ff", "c": "db_password = \"super_secret_password\"\n\nopenai_api_key = \"sk-xxxx\"\ndeepseek_api_key = \"dseek-xxxx\"\n\nmalware_bazaaar_api_key = \"malw-bazaaar-xxxx\"\nvirus_total_api_key = \"vt-xxxx\"\n\ndecompiler_api_key = \"decomp-xxxx\"\n"}, "6c66ab17dd5d4bac01bab166787a643c4e584b90": {"t": 1785750642.1860275, "n": 177, "h": "428e8f296a346f47", "c": "db_password = \"super_secret_password\"\n\nopenai_api_key = \"sk-xxxx\"\ndeepseek_api_key = \"dseek-xxxx\"\n\nmalware_bazaaar_api_key = \"malw-bazaaar-xxxx\"\nvirus_total_api_key = \"vt-xxxx\"\n"}, "db798478e1e1dd6a7dee0330a07f5a85927a0d76": {"t": 1785750643.270572, "n": 175, "h": "8a710cde99e41453", "c": "db_password = \"super_secret_password\"\nopenai_api_key = \"sk-xxxx\"\ndeepseek_api_key = \"dseek-xxxx\"\nmalware_bazaaar_api_key = \"malw-bazaaar-xxxx\"\nvirus_total_api_key = \"vt-xxxx\"\n"}, "f6511f78d99cfb1a006f983f1bbdd69823f9d66c": {"t": 1785750645.3685486, "n": 107, "h": "2e5edf7e72d38793", "c": "[default]\ndb_password = \"super_secret_password\"\nopenai_api_key = \"sk-xxxx\"\ndeepseek_api_key = \"dseek-xxxx\"\n"}}, "config/settings.toml": {"95c6933ae08544d6b66a2e9567ff7ea1e68a9ec7": {"t": 1785745368.2252982, "n": 1843, "h": "681ce357d698f63b", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n[collection]\nmalware_bazaar_api_key = \"@get malware_bazaar_api_key\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[baseline]\nvirus_total_url = \"https://www.virustotal.com/api/v3/files\"\nvirus_total_api_key = \"@get virus_total_api_key\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 256\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "d06d60634294a114b9d4614cdb459518029152fa": {"t": 1785750642.414558, "n": 1843, "h": "681ce357d698f63b", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n[collection]\nmalware_bazaar_api_key = \"@get malware_bazaar_api_key\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[baseline]\nvirus_total_url = \"https://www.virustotal.com/api/v3/files\"\nvirus_total_api_key = \"@get virus_total_api_key\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 256\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "d92f46c6dc135d1638ec3ca62605cb74a4dbb6ac": {"t": 1785750643.81951, "n": 1843, "h": "5d2dabb1f1f6e58f", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n[collection]\nmalware_bazaar_api_key = \"@get malware_bazaar_api_key\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[baseline]\nvirus_total_url = \"https://www.virustotal.com/api/v3/files\"\nvirus_total_api_key = \"@get virus_total_api_key\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "90cd0f49fb87234cd60f3889a039dbd37379df4b": {"t": 1785750644.892477, "n": 1789, "h": "07052f4dd80923de", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[baseline]\nvirus_total_url = \"https://www.virustotal.com/api/v3/files\"\nvirus_total_api_key = \"@get virus_total_api_key\"\n\n[decompiler]\n\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "4b7f76a75d175cddd7622fa13be1eaa38d98c249": {"t": 1785750645.9997272, "n": 1788, "h": "00f4658c86de9e7b", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[baseline]\nvirus_total_url = \"https://www.virustotal.com/api/v3/files\"\nvirus_total_api_key = \"@get virustotal_api_key\"\n\n[decompiler]\n\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "a567421c91a0636f64f804fb154049dec4e07a11": {"t": 1785750647.0978463, "n": 1668, "h": "389d4f2014d0ca96", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[decompiler]\n\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "dfad596a6f712ba6be6bb70825251d826c0fd163": {"t": 1785750648.1561272, "n": 1755, "h": "0a68d43b9e7f0368", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n[decompiler]\nenabled = true\nghidra_url = \"http://localhost:8000\"\napi_key = \"@get decompiler_api_key\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "611253d48a9cb06ec000faa650dafa37a8c6b7ef": {"t": 1785750649.2517877, "n": 1653, "h": "b72401dd9af97d16", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\ndefault_user_agent = \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "9dab7c442ebed6a5402a0f8cb2b52f25277b8e7f": {"t": 1785750650.3398676, "n": 1528, "h": "0435a1784f741733", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\nmalpedia_base_url = \"https://malpedia.caad.fkie.fraunhofer.de\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "a4afb05a6331041d5b925221029f7cb7cae51183": {"t": 1785750651.413616, "n": 1465, "h": "132739184048b893", "c": "\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n# Tagger LLM settings\n[tagger]\nmodel = \"gpt-4.1-mini\"\napi_key = \"@get openai_api_key\"\nsystem_prompt = \"You are a malware analysis assistant. Your task is to extract 3-7 highly relevant tags from the provided text. Tags should be keywords, entities, malware families, techniques, countries, CVEs, or other important topics.\\n\\nReturn ONLY a valid Python list of strings, with no explanation or extra text. Example: ['APT28', 'phishing', 'Russia', 'CVE-2023-38831', 'lateral movement'].\\n\\nText to tag:\"\ntemperature = 0.2\nmax_tokens = 128\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "921f8aace33520b3321f48867e8d45837a88ccc1": {"t": 1785750652.476561, "n": 926, "h": "eed4bbb180a80012", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "9fff2528448af5f842bbe2c6b891e1540907cb37": {"t": 1785750653.5638719, "n": 1134, "h": "5326bb3fd395ed9a", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Ingestion settings\n[ingestion]\n[ingestion.vxu]\nbase_url = \"https://vx-underground.org\"\ndownload_dir = \"./data/vx_underground\"\nuser_agent = \"rexis-ingestor/1.0 (+https://github.com/andremmfaria/rexis)\"\nreq_timeout = 30\nrate_delay_sec = 0.5\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "24655b35a9270e6a43be8b4e7946f62136218b39": {"t": 1785750654.646919, "n": 892, "h": "f39864ebd5f6ff94", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", embedding_model_limit=8192, query_model = \"deepseek-chat\"}\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "11b01146f7bacc825a154f167240fb197a48e653": {"t": 1785750655.7489455, "n": 836, "h": "2b6a8b8ea82b6a81", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", query_model = \"deepseek-chat\"}\n\n[pipeline_params]\nretriever_top_k = 5\nprompt_template = \"\"\"\nYou are a malware analysis assistant. Based on the following malware metadata, answer the user's question:\n\n{% for doc in documents %}\nDocument {{ loop.index }}:\n{{ doc.content }}\n\n{% endfor %}\n\nUser Question:\n{{ query }}\n\nAnswer:\n\"\"\"\n"}, "6c66ab17dd5d4bac01bab166787a643c4e584b90": {"t": 1785750656.8081036, "n": 579, "h": "a016e81d0b7886db", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\n[db]\nhost = \"localhost\"\nport = 5432\nname = \"rexis\"\nuser = \"postgres\"\npassword = \"@get db_password\"\n\n# Model metadata\n[models]\nopenai = {base_url = \"https://api.openai.com/v1\", api_key = \"@get openai_api_key\", embedding_model = \"text-embedding-3-small\", query_model = \"gpt-4o-2024-08-06\"}\ndeepseek = {base_url = \"https://api.deepseek.com\", api_key = \"@get deepseek_api_key\", embedding_model = \"text-embedding-3-small\", query_model = \"deepseek-chat\"}\n\n\n# RAG parameters\nretriever_top_k = 5\n"}, "db798478e1e1dd6a7dee0330a07f5a85927a0d76": {"t": 1785750657.8656976, "n": 352, "h": "8ad4ee63adb75c8f", "c": "# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\ndb_host = \"localhost\"\ndb_port = 5432\ndb_name = \"rexis\"\ndb_user = \"postgres\"\ndb_password = \"@db_password\"\n\n# Embedding model defaults\nembedding_provider = \"openai\" # or \"deepseek\"\nembedding_model = \"text-embedding-3-small\"\n\n# RAG parameters\nretriever_top_k = 5\n"}, "f6511f78d99cfb1a006f983f1bbdd69823f9d66c": {"t": 1785750658.9201128, "n": 362, "h": "53bacf09c146f737", "c": "[default]\n# General project settings\nproject_name = \"REXIS\"\nlog_level = \"INFO\"\n\n# Vector DB settings\ndb_host = \"localhost\"\ndb_port = 5432\ndb_name = \"rexis\"\ndb_user = \"postgres\"\ndb_password = \"@db_password\"\n\n# Embedding model defaults\nembedding_provider = \"openai\" # or \"deepseek\"\nembedding_model = \"text-embedding-3-small\"\n\n# RAG parameters\nretriever_top_k = 5\n"}}}} |