Claim TraceabilityEvidence Graph

Security Governance Claim Traceability

Top conclusions linked to resource cards, URLs, evidence quality and blocker constraints.

Workbench EvidenceMarkdown source

Claims

Each top conclusion with evidence refs and blockers.

L2_direct_public_evidence

MOLOT 已可作为安全专题的直接论文主线，但它证明的是 Python/JavaScript 包 SAST 静态恶意代码检测，不证明 FCD C/C++ 表。

Policy: blocked_or_boundary_claim

Evidence refs: molot_system_card

URLs: https://arxiv.org/abs/2606.07792

Blockers: block_fcd_pdg_table_upgrade, block_open_malicious_code_bench_artifact_upgrade, block_agent_gate_performance_upgrade

L2_multi_paper_supported_principle

行为链检测已经从方法背景上移为安全专题的核心原则：单个 API/关键词不足以判断恶意目的，必须看 activity chain。

Policy: public_evidence_supported

Evidence refs: molot_system_card, cerebro_behavior_sequence, npm_malicious_detection_benchmark

URLs: https://arxiv.org/abs/2606.07792, https://arxiv.org/abs/2309.02637, https://arxiv.org/abs/2603.27549

Blockers: none

L2_direct_operational_constraint

安全门禁不能只追求召回率；误报预算和 block/ask_human 分流必须前置。

Policy: blocked_or_boundary_claim

Evidence refs: python_malware_benchmark, npm_malicious_detection_benchmark, molot_system_card

URLs: https://arxiv.org/abs/2209.13288, https://arxiv.org/abs/2603.27549, https://arxiv.org/abs/2606.07792

Blockers: block_agent_gate_performance_upgrade

L2_direct_plus_foundation_evidence

SHAP/解释不应停留在可视化，而应上移为 reviewer evidence：source_locations、activity_chain、reviewer_action 必须绑定。

Policy: background_with_overclaim_risk

Evidence refs: molot_system_card, shap_unified_approach

URLs: https://arxiv.org/abs/2606.07792, https://arxiv.org/abs/1705.07874

Blockers: none

L1_background_evidence_with_boundary

PDG/Graph 证据可作为源码/片段相似检测的第二检测面，但不能上移为 MOLOT 性能或 FCD 表直接证据。

Policy: blocked_or_boundary_claim

Evidence refs: pdg_holmes_graph_siamese, source_code_is_graph, fcd_pdg_performance_unresolved

URLs: https://arxiv.org/abs/2011.11228, https://arxiv.org/abs/2312.16488, user_provided_json

Blockers: block_fcd_pdg_table_upgrade

L0_to_L1_design_candidate

LRI Agent Gate 已具备合同、schema 和 MVP 用例设计，但还不能上移为已验证性能结论。

Policy: blocked_or_boundary_claim

Evidence refs: molot_system_card, python_malware_benchmark

URLs: https://arxiv.org/abs/2606.07792, https://arxiv.org/abs/2209.13288

Blockers: block_agent_gate_performance_upgrade

Edges

Claim → evidence and claim → blocker relations.

FromRelationTo
rc_01_molot_is_direct_mainlinesupported_bymolot_system_card
rc_01_molot_is_direct_mainlineblocked_byblock_fcd_pdg_table_upgrade
rc_01_molot_is_direct_mainlineblocked_byblock_open_malicious_code_bench_artifact_upgrade
rc_01_molot_is_direct_mainlineblocked_byblock_agent_gate_performance_upgrade
rc_02_behavior_chain_becomes_detection_principlesupported_bymolot_system_card
rc_02_behavior_chain_becomes_detection_principlesupported_bycerebro_behavior_sequence
rc_02_behavior_chain_becomes_detection_principlesupported_bynpm_malicious_detection_benchmark
rc_03_false_positive_budget_is_requiredsupported_bypython_malware_benchmark
rc_03_false_positive_budget_is_requiredsupported_bynpm_malicious_detection_benchmark
rc_03_false_positive_budget_is_requiredsupported_bymolot_system_card
rc_03_false_positive_budget_is_requiredblocked_byblock_agent_gate_performance_upgrade
rc_04_explainability_is_review_evidencesupported_bymolot_system_card
rc_04_explainability_is_review_evidencesupported_byshap_unified_approach
rc_05_graph_pdg_is_secondary_chain_not_molot_metricsupported_bypdg_holmes_graph_siamese
rc_05_graph_pdg_is_secondary_chain_not_molot_metricsupported_bysource_code_is_graph
rc_05_graph_pdg_is_secondary_chain_not_molot_metricsupported_byfcd_pdg_performance_unresolved
rc_05_graph_pdg_is_secondary_chain_not_molot_metricblocked_byblock_fcd_pdg_table_upgrade
rc_06_agent_gate_is_design_candidate_not_verified_resultsupported_bymolot_system_card
rc_06_agent_gate_is_design_candidate_not_verified_resultsupported_bypython_malware_benchmark
rc_06_agent_gate_is_design_candidate_not_verified_resultblocked_byblock_agent_gate_performance_upgrade