{ "schema": "lri.security-public-evidence-snapshots.v1", "snapshot_count": 10, "snapshots": [ { "id": "pub_molot_arxiv_snapshot", "source_url": "https://arxiv.org/abs/2606.07792", "source_title": "MOLOT System Card: Malicious Operational Logic Observation Transformer", "resource_card_id": "molot_system_card", "direct_claims": [ "MOLOT is a static malicious-code detection system designed for SAST settings.", "MOLOT represents source code as behavior sequences derived from static call graphs.", "MOLOT includes an explanation stage that maps suspicious behavior activities to source-code locations.", "MOLOT is evaluated on Python and JavaScript packages and compared with open-source tools.", "The paper states Open Malicious-Code Bench is released for reproducible evaluation." ], "chain_use": "LRI 可把它作为 MOLOT 检测主线的直接论文证据。", "not_enough_for": [ "FCD C/C++ table", "PDG binary clone detection metrics", "independent Open Malicious-Code Bench repository URL" ], "upgrade_action": "抽取论文表格、benchmark 地址和限制章节；若找到独立 benchmark URL，再把 benchmark 从 paper-stated 升级为 public_artifact." }, { "id": "pub_cerebro_arxiv_snapshot", "source_url": "https://arxiv.org/abs/2309.02637", "source_title": "Malicious Package Detection in NPM and PyPI using a Single Model of Malicious Behavior Sequence", "resource_card_id": "cerebro_behavior_sequence", "direct_claims": [ "Cerebro detects malicious packages across NPM and PyPI.", "It curates high-level malicious behavior abstractions and organizes features into behavior sequences.", "It fine-tunes BERT to model malicious behavior semantics.", "It reported finding new malicious packages in PyPI and NPM." ], "chain_use": "支撑“行为序列 + BERT”是恶意包检测的公开方法线，作为 MOLOT 前序/旁证。", "not_enough_for": [ "MOLOT performance", "Agent Gate execution correctness", "PDG/FCD metrics" ], "upgrade_action": "抽取行为 abstraction taxonomy，映射到 LRI malicious_activity_taxonomy." }, { "id": "pub_pypi_benchmark_snapshot", "source_url": "https://arxiv.org/abs/2209.13288", "source_title": "A Benchmark Comparison of Python Malware Detection Approaches", "resource_card_id": "python_malware_benchmark", "direct_claims": [ "The paper builds a benchmark dataset for Python/PyPI malware detection.", "It compares PyPI checks, Bandit4Mal, and OSSGadget OSS Detect Backdoor.", "It reports repository administrators consider 0.01% false-positive rate too high in PyPI moderation.", "Measured tools show high false-positive rates in the evaluated setting." ], "chain_use": "支撑 LRI Gate 必须将 block 与 ask_human 分离，并为误报建立 budget。", "not_enough_for": [ "NPM ecosystem performance", "MOLOT benchmark numbers", "binary download scanning" ], "upgrade_action": "抽取具体 baseline 表，建立 false_positive_budget 字段。" }, { "id": "pub_npm_benchmark_snapshot", "source_url": "https://arxiv.org/abs/2603.27549", "source_title": "Understanding NPM Malicious Package Detection: A Benchmark-Driven Empirical Analysis", "resource_card_id": "npm_malicious_detection_benchmark", "direct_claims": [ "The paper builds a dataset with malicious and benign NPM packages.", "It annotates 11 behavior categories and 8 evasion techniques.", "It evaluates 8 tools across 13 variants.", "It states a behavioral chain disambiguates malicious purpose better than a single API call.", "It reports strategic tool combinations can achieve high accuracy and F1." ], "chain_use": "支撑 LRI 对 JS/TS Skills 插件采用 activity chain 而非 isolated API signal。", "not_enough_for": [ "Python-only statement taxonomy", "MOLOT direct method", "FCD/PDG table" ], "upgrade_action": "抽取 11 behavior categories 和 8 evasion techniques，映射 LRI activity taxonomy." }, { "id": "pub_oss_malicious_packages_snapshot", "source_url": "https://arxiv.org/abs/2404.04991", "source_title": "OSS Malicious Package Analysis in the Wild", "resource_card_id": "oss_malicious_package_analysis", "direct_claims": [ "The paper curates 23,425 malicious packages from scattered online sources.", "It proposes a knowledge graph representation of OSS malware.", "It notes low overlap between different online sources.", "It describes a lifecycle of changing, release, detection, and removal." ], "chain_use": "支撑 LRI 建多源 evidence registry，而非单一 benchmark 依赖。", "not_enough_for": [ "single detector precision", "Agent Gate runtime metrics" ], "upgrade_action": "增加 campaign_context、variant_family、source_overlap 字段候选。" }, { "id": "pub_backstabbers_snapshot", "source_url": "https://arxiv.org/abs/2005.09535", "source_title": "Backstabber’s Knife Collection", "resource_card_id": "backstabbers_knife_collection", "direct_claims": [ "The paper reviews OSS supply-chain attacks involving malicious code injection into packages.", "It presents 174 malicious packages from npm, PyPI, and RubyGems.", "It proposes attack trees for dependency-tree injection and execution timing." ], "chain_use": "支撑 LRI 对 install-time/import-time/runtime 执行阶段分开建模。", "not_enough_for": [ "2026 ecosystem coverage", "MOLOT numbers", "MCP-specific tool attacks" ], "upgrade_action": "把 attack tree 映射到 artifact_threat_taxonomy 的 execution_phase 字段。" }, { "id": "pub_statement_taxonomy_snapshot", "source_url": "https://arxiv.org/abs/2512.12559", "source_title": "Unveiling Malicious Logic", "resource_card_id": "unveiling_malicious_logic", "direct_claims": [ "The paper builds a statement-level dataset of malicious Python packages.", "It labels malicious indicator occurrences at code-statement level.", "It derives 47 malicious indicators across 7 types.", "It applies sequential pattern mining to recurring indicator sequences." ], "chain_use": "支撑 LRI 将模型定位结果转化为 statement-level indicator 和 reviewer checklist。", "not_enough_for": [ "JavaScript/NPM coverage", "binary clone detection" ], "upgrade_action": "抽取 indicator taxonomy，补 reviewer_evidence_schema 的 indicator_type。" }, { "id": "pub_llm_pypi_snapshot", "source_url": "https://arxiv.org/abs/2504.13769", "source_title": "Detecting Malicious Source Code in PyPI Packages with LLMs: Does RAG Come in Handy?", "resource_card_id": "llm_pypi_malicious_detection", "direct_claims": [ "The paper evaluates LLMs, RAG, and few-shot learning for malicious PyPI source-code detection.", "It integrates YARA rules, GitHub Security Advisories, and malicious code snippets.", "It reports RAG performed modestly while few-shot learning was more effective in the evaluation." ], "chain_use": "支撑 LRI 将 LLM 作为 labeling/review assistant，而非唯一自动 block 信号。", "not_enough_for": [ "MOLOT direct performance", "PDG/FCD metrics" ], "upgrade_action": "建立 LLM-review 与 static evidence 的 disagreement workflow。" }, { "id": "pub_shap_snapshot", "source_url": "https://arxiv.org/abs/1705.07874", "source_title": "A Unified Approach to Interpreting Model Predictions", "resource_card_id": "shap_unified_approach", "direct_claims": [ "The paper introduces SHAP as a unified framework for interpreting model predictions.", "SHAP assigns each feature an importance value for a particular prediction.", "It connects additive feature importance methods with theoretical properties." ], "chain_use": "支撑 explanation_tokens/activity attribution 字段，但不能单独决定安全动作。", "not_enough_for": [ "malicious code localization without model-specific mapping", "block decision by itself" ], "upgrade_action": "将 SHAP 输出绑定 source_locations 和 reviewer_actions。" }, { "id": "unresolved_fcd_search_snapshot", "source_url": "user_provided_json", "source_title": "FCD / PDG C/C++ performance table search result", "resource_card_id": "fcd_pdg_performance_unresolved", "direct_claims": [ "Queries for the exact c-ares/jasper/openssl/rsync FCD speed table did not return a public original source in this iteration.", "The table remains only a user-provided lead." ], "chain_use": "阻止将 FCD 表升级为 direct evidence。", "not_enough_for": [ "any verified precision/recall claim", "association with MOLOT or PDG papers" ], "upgrade_action": "继续用 exact project/version/table terms 搜索原始论文、报告或 slide。" } ] }