briefloop 0.11.12__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- briefloop-0.11.12.dist-info/METADATA +473 -0
- briefloop-0.11.12.dist-info/RECORD +424 -0
- briefloop-0.11.12.dist-info/WHEEL +5 -0
- briefloop-0.11.12.dist-info/entry_points.txt +3 -0
- briefloop-0.11.12.dist-info/licenses/LICENSE +22 -0
- briefloop-0.11.12.dist-info/top_level.txt +1 -0
- multi_agent_brief/__init__.py +36 -0
- multi_agent_brief/analysis_blocks/__init__.py +11 -0
- multi_agent_brief/analysis_blocks/builder.py +206 -0
- multi_agent_brief/analysis_blocks/renderer.py +237 -0
- multi_agent_brief/analysis_blocks/schemas.py +52 -0
- multi_agent_brief/analysis_modules/__init__.py +8 -0
- multi_agent_brief/analysis_modules/base.py +87 -0
- multi_agent_brief/analysis_modules/market_competitor/__init__.py +68 -0
- multi_agent_brief/analysis_modules/market_competitor/auditor.py +290 -0
- multi_agent_brief/analysis_modules/market_competitor/config.py +187 -0
- multi_agent_brief/analysis_modules/market_competitor/event_builder.py +226 -0
- multi_agent_brief/analysis_modules/market_competitor/renderer.py +199 -0
- multi_agent_brief/analysis_modules/market_competitor/schemas.py +518 -0
- multi_agent_brief/analysis_modules/policy_regulatory/__init__.py +7 -0
- multi_agent_brief/analysis_modules/policy_regulatory/audit.py +195 -0
- multi_agent_brief/analysis_modules/policy_regulatory/module.py +387 -0
- multi_agent_brief/analysis_modules/policy_regulatory/schemas.py +129 -0
- multi_agent_brief/analysis_modules/registry.py +73 -0
- multi_agent_brief/audience/__init__.py +19 -0
- multi_agent_brief/audience/profiles.py +246 -0
- multi_agent_brief/audience_memory/__init__.py +23 -0
- multi_agent_brief/audience_memory/profile.py +358 -0
- multi_agent_brief/audit/__init__.py +36 -0
- multi_agent_brief/audit/case_applicability.py +181 -0
- multi_agent_brief/audit/deterministic.py +471 -0
- multi_agent_brief/audit/editorial_governance.py +404 -0
- multi_agent_brief/audit/final_quality.py +764 -0
- multi_agent_brief/audit/harness.py +242 -0
- multi_agent_brief/audit/interfaces.py +96 -0
- multi_agent_brief/audit/limitation_hygiene.py +238 -0
- multi_agent_brief/audit/proposal_boundary.py +34 -0
- multi_agent_brief/audit/redaction.py +27 -0
- multi_agent_brief/audit/rule_packs.py +96 -0
- multi_agent_brief/audit/semantic.py +435 -0
- multi_agent_brief/capabilities/__init__.py +22 -0
- multi_agent_brief/capabilities/catalog.py +215 -0
- multi_agent_brief/capabilities/detect.py +199 -0
- multi_agent_brief/capabilities/models.py +60 -0
- multi_agent_brief/capabilities/recommend.py +206 -0
- multi_agent_brief/cli/__init__.py +2 -0
- multi_agent_brief/cli/analysis_commands.py +140 -0
- multi_agent_brief/cli/approval_commands.py +107 -0
- multi_agent_brief/cli/audit_commands.py +59 -0
- multi_agent_brief/cli/capability_commands.py +548 -0
- multi_agent_brief/cli/claude_commands.py +159 -0
- multi_agent_brief/cli/competitors_commands.py +182 -0
- multi_agent_brief/cli/controls_commands.py +137 -0
- multi_agent_brief/cli/deliver_commands.py +942 -0
- multi_agent_brief/cli/eval_cases_commands.py +99 -0
- multi_agent_brief/cli/experiments_commands.py +395 -0
- multi_agent_brief/cli/feedback_commands.py +192 -0
- multi_agent_brief/cli/finalize_commands.py +153 -0
- multi_agent_brief/cli/gates_commands.py +174 -0
- multi_agent_brief/cli/hermes_commands.py +356 -0
- multi_agent_brief/cli/improve_commands.py +224 -0
- multi_agent_brief/cli/init_commands.py +564 -0
- multi_agent_brief/cli/init_wizard.py +1473 -0
- multi_agent_brief/cli/input_commands.py +237 -0
- multi_agent_brief/cli/main.py +288 -0
- multi_agent_brief/cli/onboard_commands.py +193 -0
- multi_agent_brief/cli/product_commands.py +1449 -0
- multi_agent_brief/cli/provenance_commands.py +100 -0
- multi_agent_brief/cli/release_commands.py +68 -0
- multi_agent_brief/cli/repair_commands.py +169 -0
- multi_agent_brief/cli/run_commands.py +292 -0
- multi_agent_brief/cli/runtime_commands.py +81 -0
- multi_agent_brief/cli/secrets_commands.py +232 -0
- multi_agent_brief/cli/semantic_support_commands.py +80 -0
- multi_agent_brief/cli/sources_commands.py +647 -0
- multi_agent_brief/cli/start_commands.py +8 -0
- multi_agent_brief/cli/state_commands.py +446 -0
- multi_agent_brief/cli/status_commands.py +26 -0
- multi_agent_brief/cli/workbuddy_commands.py +103 -0
- multi_agent_brief/configs/artifact_contracts.yaml +514 -0
- multi_agent_brief/configs/orchestrator_contract.yaml +181 -0
- multi_agent_brief/configs/policy_packs/default.yaml +65 -0
- multi_agent_brief/configs/policy_profiles/evidence_extract_default.yaml +58 -0
- multi_agent_brief/configs/policy_profiles/finance_default.yaml +44 -0
- multi_agent_brief/configs/policy_profiles/internet_default.yaml +44 -0
- multi_agent_brief/configs/policy_profiles/manufacturing_default.yaml +33 -0
- multi_agent_brief/configs/policy_profiles/solar_manufacturing_default.yaml +64 -0
- multi_agent_brief/configs/report_packs/evidence_extract.yaml +48 -0
- multi_agent_brief/configs/report_packs/management_monthly.yaml +36 -0
- multi_agent_brief/configs/report_packs/market_weekly.yaml +36 -0
- multi_agent_brief/configs/report_packs/solar_industry_periodic.yaml +50 -0
- multi_agent_brief/configs/report_templates/evidence_extract.yaml +55 -0
- multi_agent_brief/configs/report_templates/management_monthly.yaml +48 -0
- multi_agent_brief/configs/report_templates/market_weekly.yaml +49 -0
- multi_agent_brief/configs/report_templates/solar_industry_periodic.yaml +75 -0
- multi_agent_brief/configs/stage_specs.yaml +180 -0
- multi_agent_brief/contracts/__init__.py +41 -0
- multi_agent_brief/contracts/base.py +99 -0
- multi_agent_brief/contracts/errors.py +41 -0
- multi_agent_brief/contracts/migrations/__init__.py +5 -0
- multi_agent_brief/contracts/migrations/claim_v1_to_v2.py +33 -0
- multi_agent_brief/contracts/registry.py +159 -0
- multi_agent_brief/contracts/role_topology.py +25 -0
- multi_agent_brief/contracts/schemas/__init__.py +35 -0
- multi_agent_brief/contracts/schemas/analysis_pack.py +135 -0
- multi_agent_brief/contracts/schemas/atomic_claim_graph.py +304 -0
- multi_agent_brief/contracts/schemas/audit_report.py +80 -0
- multi_agent_brief/contracts/schemas/candidate_item.py +51 -0
- multi_agent_brief/contracts/schemas/claim.py +197 -0
- multi_agent_brief/contracts/schemas/claim_draft.py +386 -0
- multi_agent_brief/contracts/schemas/claim_support_matrix.py +334 -0
- multi_agent_brief/contracts/schemas/evidence_span_registry.py +270 -0
- multi_agent_brief/contracts/schemas/policy_profile.py +247 -0
- multi_agent_brief/contracts/schemas/report_spec.py +257 -0
- multi_agent_brief/contracts/schemas/semantic_assessment_report.py +480 -0
- multi_agent_brief/contracts/schemas/source_evidence_pack_manifest.py +233 -0
- multi_agent_brief/contracts/schemas/source_item.py +96 -0
- multi_agent_brief/contracts/semantic_assessment_status.py +12 -0
- multi_agent_brief/contracts/source_metadata.py +252 -0
- multi_agent_brief/contracts/target_contract.py +460 -0
- multi_agent_brief/contracts/validator.py +225 -0
- multi_agent_brief/controls/__init__.py +23 -0
- multi_agent_brief/controls/contract.py +106 -0
- multi_agent_brief/controls/switchboard.py +1076 -0
- multi_agent_brief/core/__init__.py +2 -0
- multi_agent_brief/core/citations.py +14 -0
- multi_agent_brief/core/claim_ledger.py +94 -0
- multi_agent_brief/core/config.py +191 -0
- multi_agent_brief/core/env.py +69 -0
- multi_agent_brief/core/previous.py +83 -0
- multi_agent_brief/core/schemas.py +181 -0
- multi_agent_brief/core/selection.py +189 -0
- multi_agent_brief/delivery/__init__.py +6 -0
- multi_agent_brief/delivery/base.py +36 -0
- multi_agent_brief/delivery/feishu.py +204 -0
- multi_agent_brief/delivery/gws.py +237 -0
- multi_agent_brief/evaluation_cases/__init__.py +15 -0
- multi_agent_brief/evaluation_cases/contract.py +429 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/approved_guidance_materialized/workspace/config.yaml +18 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/approved_guidance_materialized/workspace/improvement/ledger.jsonl +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/approved_guidance_materialized/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/approved_guidance_materialized/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/company_event_missing_latest_official_check/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/company_event_missing_latest_official_check/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/company_event_missing_latest_official_check/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/control_switchboard_selection_is_not_execution/workspace/config.yaml +12 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/control_switchboard_selection_is_not_execution/workspace/sources.yaml +6 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/control_switchboard_selection_is_not_execution/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/feedback_triage_required/workspace/config.yaml +6 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/feedback_triage_required/workspace/input/human_feedback.md +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/feedback_triage_required/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/feedback_triage_required/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/config.yaml +8 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/brief.md +14 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/intermediate/audited_brief.md +5 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/intermediate/claim_ledger.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/final_abstract_quality_warning_surface/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/formal_release_missing_human_approval/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/formal_release_missing_human_approval/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/formal_release_missing_human_approval/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/guidance_manifestation_not_observable/workspace/config.yaml +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/guidance_manifestation_not_observable/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/guidance_manifestation_not_observable/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/media_only_legal_policy_blocks_research_review/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/media_only_legal_policy_blocks_research_review/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/media_only_legal_policy_blocks_research_review/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/mixed_metric_scope_support_blocker/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/mixed_metric_scope_support_blocker/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/mixed_metric_scope_support_blocker/workspace/user.md +5 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/config.yaml +6 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/audited_brief.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/claim_ledger.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/feedback_issues.json +25 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/repair_plan.json +27 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/planned_blocking_issue_cannot_continue/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/config.yaml +8 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/audited_brief.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/claim_ledger.json +11 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/quality_gate_report.json +22 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/provenance_projection_minimal/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/output/intermediate/audited_brief.md +5 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/output/intermediate/claim_ledger.json +40 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/sources.yaml +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_source_appendix/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/config.yaml +8 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/brief.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/intermediate/audited_brief.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/intermediate/claim_ledger.json +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reader_facing_target_relevance/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/release_readiness_forged_event_blocker/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/release_readiness_forged_event_blocker/workspace/output/intermediate/release_readiness_report.json +31 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/release_readiness_forged_event_blocker/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/release_readiness_forged_event_blocker/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/config.yaml +18 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/improvement/ledger.jsonl +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/improvement/memory.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/output/intermediate/improvement_memory_snapshot.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/reverted_entry_removed_from_next_snapshot/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/config.yaml +12 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/input/sources/source-001.txt +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/brief.md +37 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/delivery/brief.md +37 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/atomic_claim_graph.json +17 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/audit_report.json +5 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/audited_brief.md +37 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/claim_ledger.json +20 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/evidence_span_registry.json +20 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/finalize_report.json +40 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/gates/auditor_quality_gate_report.json +42 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/gates/finalize_quality_gate_report.json +42 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/intermediate/screened_candidates.json +33 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/source_appendix.md +12 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/output/source_appendix_trace.md +10 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/report_spec.yaml +24 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/sources.yaml +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/same_evidence_reader_quality_regression/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/source_evidence_pack_blocks_non_evidence_file/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/source_evidence_pack_blocks_non_evidence_file/workspace/input/sources/README.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/source_evidence_pack_blocks_non_evidence_file/workspace/output/intermediate/source_evidence_pack_manifest.json +26 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/source_evidence_pack_blocks_non_evidence_file/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/source_evidence_pack_blocks_non_evidence_file/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/config.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/output/intermediate/audited_brief.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/output/intermediate/claim_ledger.json +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/stale_current_claim/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/third_party_price_snapshot_formal_block/workspace/config.yaml +13 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/third_party_price_snapshot_formal_block/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/third_party_price_snapshot_formal_block/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/trajectory_retry_budget_exhausted/workspace/config.yaml +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/trajectory_retry_budget_exhausted/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/trajectory_retry_budget_exhausted/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unapproved_entry_not_materialized/workspace/config.yaml +18 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unapproved_entry_not_materialized/workspace/improvement/ledger.jsonl +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unapproved_entry_not_materialized/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unapproved_entry_not_materialized/workspace/user.md +4 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unauthorized_institution_branding_blocks_release/workspace/config.yaml +18 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unauthorized_institution_branding_blocks_release/workspace/sources.yaml +9 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unauthorized_institution_branding_blocks_release/workspace/user.md +5 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/config.yaml +8 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/output/intermediate/audit_report.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/output/intermediate/audited_brief.md +8 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/output/intermediate/candidate_claims.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/output/intermediate/claim_ledger.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/output/intermediate/screened_candidates.json +1 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/sources.yaml +2 -0
- multi_agent_brief/evaluation_cases/fixtures/cases/unsupported_material_fact/workspace/user.md +3 -0
- multi_agent_brief/evaluation_cases/fixtures/manifest.yaml +861 -0
- multi_agent_brief/evaluation_cases/fixtures.py +20 -0
- multi_agent_brief/evaluation_cases/runner.py +1603 -0
- multi_agent_brief/experiments/__init__.py +45 -0
- multi_agent_brief/experiments/experiment_080.py +5835 -0
- multi_agent_brief/experiments/schemas.py +63 -0
- multi_agent_brief/experiments/target_contract.py +10 -0
- multi_agent_brief/feedback/__init__.py +17 -0
- multi_agent_brief/feedback/feedback_contract.py +482 -0
- multi_agent_brief/feedback/feedback_state.py +898 -0
- multi_agent_brief/hermes/__init__.py +25 -0
- multi_agent_brief/hermes/adapter.py +832 -0
- multi_agent_brief/improvement/__init__.py +83 -0
- multi_agent_brief/improvement/contract.py +1114 -0
- multi_agent_brief/improvement/memory.py +451 -0
- multi_agent_brief/improvement/product_definition.py +228 -0
- multi_agent_brief/improvement/state.py +961 -0
- multi_agent_brief/inputs/__init__.py +2 -0
- multi_agent_brief/inputs/classifier.py +322 -0
- multi_agent_brief/inputs/contracts.py +59 -0
- multi_agent_brief/inputs/extractor.py +310 -0
- multi_agent_brief/install/__init__.py +2 -0
- multi_agent_brief/install/writer.py +97 -0
- multi_agent_brief/onboarding/__init__.py +11 -0
- multi_agent_brief/onboarding/io.py +103 -0
- multi_agent_brief/onboarding/mapper.py +484 -0
- multi_agent_brief/onboarding/schema.py +46 -0
- multi_agent_brief/orchestrator/__init__.py +2 -0
- multi_agent_brief/orchestrator/fact_layer_import.py +352 -0
- multi_agent_brief/orchestrator/handoff.py +2209 -0
- multi_agent_brief/orchestrator/role_topology.py +42 -0
- multi_agent_brief/orchestrator/run_archive.py +888 -0
- multi_agent_brief/orchestrator/run_integrity.py +346 -0
- multi_agent_brief/orchestrator/runtime_state/__init__.py +70 -0
- multi_agent_brief/orchestrator/runtime_state/_io.py +167 -0
- multi_agent_brief/orchestrator/runtime_state/_transactions.py +298 -0
- multi_agent_brief/orchestrator/runtime_state/artifact_registry.py +1825 -0
- multi_agent_brief/orchestrator/runtime_state/atomic_claim_graph.py +96 -0
- multi_agent_brief/orchestrator/runtime_state/claim_ledger_freeze.py +510 -0
- multi_agent_brief/orchestrator/runtime_state/claim_metadata_enrichment.py +836 -0
- multi_agent_brief/orchestrator/runtime_state/claim_support_matrix.py +582 -0
- multi_agent_brief/orchestrator/runtime_state/completion_gates.py +672 -0
- multi_agent_brief/orchestrator/runtime_state/contracts_loader.py +66 -0
- multi_agent_brief/orchestrator/runtime_state/decisions.py +239 -0
- multi_agent_brief/orchestrator/runtime_state/errors.py +61 -0
- multi_agent_brief/orchestrator/runtime_state/event_log.py +256 -0
- multi_agent_brief/orchestrator/runtime_state/evidence_span_registry.py +233 -0
- multi_agent_brief/orchestrator/runtime_state/fact_layer.py +755 -0
- multi_agent_brief/orchestrator/runtime_state/identity.py +90 -0
- multi_agent_brief/orchestrator/runtime_state/lifecycle.py +675 -0
- multi_agent_brief/orchestrator/runtime_state/manifest.py +82 -0
- multi_agent_brief/orchestrator/runtime_state/operations.py +224 -0
- multi_agent_brief/orchestrator/runtime_state/paths.py +38 -0
- multi_agent_brief/orchestrator/runtime_state/repair.py +984 -0
- multi_agent_brief/orchestrator/runtime_state/safety_surfaces.py +147 -0
- multi_agent_brief/orchestrator/runtime_state/semantic_assessment_report.py +692 -0
- multi_agent_brief/orchestrator/runtime_state/semantic_support_acceptance.py +683 -0
- multi_agent_brief/orchestrator/runtime_state/source_evidence_pack.py +131 -0
- multi_agent_brief/orchestrator/runtime_state/stage_completion.py +1320 -0
- multi_agent_brief/orchestrator/runtime_state/trajectory.py +200 -0
- multi_agent_brief/orchestrator/runtime_state/workflow.py +350 -0
- multi_agent_brief/orchestrator/source_evidence.py +43 -0
- multi_agent_brief/orchestrator/timing.py +410 -0
- multi_agent_brief/orchestrator_contract.py +96 -0
- multi_agent_brief/outputs/__init__.py +2 -0
- multi_agent_brief/outputs/atomic_reader_projection.py +288 -0
- multi_agent_brief/outputs/docx.py +32 -0
- multi_agent_brief/outputs/finalize.py +1036 -0
- multi_agent_brief/outputs/ib_docx.py +983 -0
- multi_agent_brief/outputs/naming.py +38 -0
- multi_agent_brief/outputs/pdf.py +20 -0
- multi_agent_brief/outputs/reader_final_gate.py +509 -0
- multi_agent_brief/outputs/reader_projection.py +397 -0
- multi_agent_brief/outputs/source_appendix.py +753 -0
- multi_agent_brief/outputs/templates/__init__.py +112 -0
- multi_agent_brief/product/__init__.py +49 -0
- multi_agent_brief/product/bundle_projection.py +686 -0
- multi_agent_brief/product/citation_profile.py +145 -0
- multi_agent_brief/product/guidance_manifestation.py +421 -0
- multi_agent_brief/product/materiality_selection.py +472 -0
- multi_agent_brief/product/policy_gate_adapter.py +116 -0
- multi_agent_brief/product/policy_profile.py +39 -0
- multi_agent_brief/product/policy_projection.py +161 -0
- multi_agent_brief/product/policy_registry.py +83 -0
- multi_agent_brief/product/policy_resolver.py +225 -0
- multi_agent_brief/product/quality_closeout.py +188 -0
- multi_agent_brief/product/quality_panel.py +2217 -0
- multi_agent_brief/product/release_approval.py +1034 -0
- multi_agent_brief/product/report_pack.py +130 -0
- multi_agent_brief/product/report_pack_aliases.py +68 -0
- multi_agent_brief/product/report_registry.py +101 -0
- multi_agent_brief/product/report_spec.py +181 -0
- multi_agent_brief/product/support_wording.py +525 -0
- multi_agent_brief/product/template_conformance.py +516 -0
- multi_agent_brief/product/template_projection.py +123 -0
- multi_agent_brief/product/template_registry.py +244 -0
- multi_agent_brief/product/template_render_plan.py +315 -0
- multi_agent_brief/product/template_renderer.py +224 -0
- multi_agent_brief/product/trajectory_regulation.py +441 -0
- multi_agent_brief/provenance/__init__.py +5 -0
- multi_agent_brief/provenance/builder.py +841 -0
- multi_agent_brief/provenance/contract.py +21 -0
- multi_agent_brief/provenance/io.py +144 -0
- multi_agent_brief/provenance/model.py +110 -0
- multi_agent_brief/provenance/references.py +43 -0
- multi_agent_brief/provenance/validator.py +135 -0
- multi_agent_brief/quality_gates/__init__.py +19 -0
- multi_agent_brief/quality_gates/contract.py +581 -0
- multi_agent_brief/quality_gates/state.py +2498 -0
- multi_agent_brief/repair/__init__.py +5 -0
- multi_agent_brief/repair/router.py +842 -0
- multi_agent_brief/runtime_assets.py +431 -0
- multi_agent_brief/sources/__init__.py +16 -0
- multi_agent_brief/sources/api_filings.py +236 -0
- multi_agent_brief/sources/api_news.py +174 -0
- multi_agent_brief/sources/base.py +142 -0
- multi_agent_brief/sources/cached_package.py +119 -0
- multi_agent_brief/sources/cli_provider.py +263 -0
- multi_agent_brief/sources/coverage.py +336 -0
- multi_agent_brief/sources/decider.py +756 -0
- multi_agent_brief/sources/doctor.py +373 -0
- multi_agent_brief/sources/evidence_pack.py +364 -0
- multi_agent_brief/sources/feishu_provider.py +326 -0
- multi_agent_brief/sources/filing_resolver.py +400 -0
- multi_agent_brief/sources/industry_packs.py +130 -0
- multi_agent_brief/sources/join.py +206 -0
- multi_agent_brief/sources/local_signal.py +81 -0
- multi_agent_brief/sources/local_signal_planner.py +636 -0
- multi_agent_brief/sources/manual.py +259 -0
- multi_agent_brief/sources/mcp_provider.py +309 -0
- multi_agent_brief/sources/mineru_provider.py +601 -0
- multi_agent_brief/sources/normalizer.py +95 -0
- multi_agent_brief/sources/opencli_provider.py +282 -0
- multi_agent_brief/sources/planner.py +178 -0
- multi_agent_brief/sources/registry.py +367 -0
- multi_agent_brief/sources/rss.py +190 -0
- multi_agent_brief/sources/search_backends/__init__.py +31 -0
- multi_agent_brief/sources/search_backends/base.py +51 -0
- multi_agent_brief/sources/search_backends/brave.py +192 -0
- multi_agent_brief/sources/search_backends/capabilities.py +94 -0
- multi_agent_brief/sources/search_backends/exa.py +180 -0
- multi_agent_brief/sources/search_backends/firecrawl.py +164 -0
- multi_agent_brief/sources/search_backends/serper.py +241 -0
- multi_agent_brief/sources/search_backends/tavily.py +124 -0
- multi_agent_brief/sources/sourcehub.py +505 -0
- multi_agent_brief/sources/web_search.py +267 -0
- multi_agent_brief/status.py +966 -0
- multi_agent_brief/tools/__init__.py +3 -0
- multi_agent_brief/tools/draft_cleanup.py +198 -0
- multi_agent_brief/workbuddy/__init__.py +1 -0
- multi_agent_brief/workbuddy/diagnose.py +569 -0
- multi_agent_brief/workbuddy/skill_pack.py +309 -0
- multi_agent_brief/workspace/__init__.py +1 -0
- multi_agent_brief/workspace/init_profile.py +45 -0
|
@@ -0,0 +1,753 @@
|
|
|
1
|
+
from __future__ import annotations
|
|
2
|
+
|
|
3
|
+
import hashlib
|
|
4
|
+
import json
|
|
5
|
+
import re
|
|
6
|
+
from dataclasses import asdict, dataclass, field
|
|
7
|
+
from pathlib import Path, PureWindowsPath
|
|
8
|
+
from typing import Any
|
|
9
|
+
from urllib.parse import urlparse
|
|
10
|
+
|
|
11
|
+
from multi_agent_brief.contracts.source_metadata import (
|
|
12
|
+
VALID_RETRIEVAL_SOURCE_TYPES,
|
|
13
|
+
VALID_SOURCE_CATEGORIES,
|
|
14
|
+
VALID_UNDERLYING_EVIDENCE_TYPES,
|
|
15
|
+
)
|
|
16
|
+
from multi_agent_brief.contracts.schemas.evidence_span_registry import EvidenceSpanRegistryContract
|
|
17
|
+
from multi_agent_brief.core.claim_ledger import ClaimLedger
|
|
18
|
+
from multi_agent_brief.core.schemas import Claim
|
|
19
|
+
|
|
20
|
+
|
|
21
|
+
_SRC_REF_RE = re.compile(r"\[src:([^\]]+)\]")
|
|
22
|
+
_LOCAL_PATH_MARKERS = ("/Users/", "/home/", "/private/", "/var/folders/")
|
|
23
|
+
_WINDOWS_USER_RE = re.compile(r"[A-Za-z]:\\Users\\")
|
|
24
|
+
_INTERNAL_ID_RE = re.compile(
|
|
25
|
+
r"\b(?:SYN_)?(?:CLAIM|SRC|SOURCE|CLM)_[A-Z0-9][A-Z0-9_-]*\b"
|
|
26
|
+
)
|
|
27
|
+
_TRACE_EXCERPT_LIMIT = 500
|
|
28
|
+
_SOURCE_CATEGORY_LABELS = {
|
|
29
|
+
"clin" + "ical_registry": "Clinical registry",
|
|
30
|
+
"company_press_release": "Company press release",
|
|
31
|
+
"industry_database": "Industry database",
|
|
32
|
+
"market_report": "Market report",
|
|
33
|
+
"news_media": "News media",
|
|
34
|
+
"other": "Other",
|
|
35
|
+
"peer_reviewed_paper": "Peer-reviewed paper",
|
|
36
|
+
"preprint": "Preprint",
|
|
37
|
+
"regulator": "Regulator",
|
|
38
|
+
}
|
|
39
|
+
_RETRIEVAL_SOURCE_TYPE_LABELS = {
|
|
40
|
+
"blog": "Blog",
|
|
41
|
+
"company_pr": "Company PR",
|
|
42
|
+
"dataset": "Dataset",
|
|
43
|
+
"filing": "Filing",
|
|
44
|
+
"local_file": "Local file",
|
|
45
|
+
"news_media": "News media",
|
|
46
|
+
"other": "Other",
|
|
47
|
+
"paper_page": "Paper page",
|
|
48
|
+
"podcast": "Podcast",
|
|
49
|
+
"social": "Social",
|
|
50
|
+
"video": "Video",
|
|
51
|
+
}
|
|
52
|
+
_UNDERLYING_EVIDENCE_TYPE_LABELS = {
|
|
53
|
+
"company_claim": "Company claim",
|
|
54
|
+
"conference_demo": "Conference demo",
|
|
55
|
+
"filing": "Filing",
|
|
56
|
+
"interview": "Interview",
|
|
57
|
+
"market_data": "Market data",
|
|
58
|
+
"media_report": "Media report",
|
|
59
|
+
"peer_reviewed_paper": "Peer-reviewed paper",
|
|
60
|
+
"regulator_record": "Regulator record",
|
|
61
|
+
"unknown": "Unknown",
|
|
62
|
+
}
|
|
63
|
+
|
|
64
|
+
|
|
65
|
+
@dataclass
|
|
66
|
+
class SourceAppendixRecord:
|
|
67
|
+
label: str
|
|
68
|
+
title: str
|
|
69
|
+
source_id: str = ""
|
|
70
|
+
source_ids: list[str] = field(default_factory=list)
|
|
71
|
+
publisher: str = ""
|
|
72
|
+
published_at: str = ""
|
|
73
|
+
retrieved_at: str = ""
|
|
74
|
+
url: str = ""
|
|
75
|
+
source_type: str = ""
|
|
76
|
+
source_category: str = ""
|
|
77
|
+
retrieval_source_type: str = ""
|
|
78
|
+
underlying_evidence_type: str = ""
|
|
79
|
+
claim_count: int = 0
|
|
80
|
+
claim_ids: list[str] = field(default_factory=list)
|
|
81
|
+
claim_ids_by_source_id: dict[str, list[str]] = field(default_factory=dict)
|
|
82
|
+
metadata_warnings: list[str] = field(default_factory=list)
|
|
83
|
+
span_count: int = 0
|
|
84
|
+
span_roles: list[str] = field(default_factory=list)
|
|
85
|
+
|
|
86
|
+
def to_dict(self) -> dict[str, Any]:
|
|
87
|
+
return asdict(self)
|
|
88
|
+
|
|
89
|
+
|
|
90
|
+
@dataclass
|
|
91
|
+
class SourceAppendixResult:
|
|
92
|
+
status: str
|
|
93
|
+
source_count: int = 0
|
|
94
|
+
cited_claim_count: int = 0
|
|
95
|
+
resolved_claim_count: int = 0
|
|
96
|
+
warnings: list[str] = field(default_factory=list)
|
|
97
|
+
markdown: str = ""
|
|
98
|
+
records: list[SourceAppendixRecord] = field(default_factory=list)
|
|
99
|
+
citation_labels: dict[str, str] = field(default_factory=dict)
|
|
100
|
+
claim_source_map: dict[str, dict[str, str]] = field(default_factory=dict)
|
|
101
|
+
trace_status: str = "not_available"
|
|
102
|
+
trace_markdown: str = ""
|
|
103
|
+
trace_source_count: int = 0
|
|
104
|
+
trace_span_count: int = 0
|
|
105
|
+
trace_warnings: list[str] = field(default_factory=list)
|
|
106
|
+
|
|
107
|
+
def to_report_fields(
|
|
108
|
+
self,
|
|
109
|
+
*,
|
|
110
|
+
source_appendix: str,
|
|
111
|
+
requested_by: str,
|
|
112
|
+
mode: str,
|
|
113
|
+
) -> dict[str, Any]:
|
|
114
|
+
return {
|
|
115
|
+
"source_appendix": source_appendix,
|
|
116
|
+
"source_appendix_generation": self.status,
|
|
117
|
+
"source_appendix_requested_by": requested_by,
|
|
118
|
+
"source_appendix_mode": mode,
|
|
119
|
+
"source_appendix_source_count": self.source_count,
|
|
120
|
+
"source_appendix_cited_claim_count": self.cited_claim_count,
|
|
121
|
+
"source_appendix_resolved_claim_count": self.resolved_claim_count,
|
|
122
|
+
"source_appendix_warnings": list(self.warnings),
|
|
123
|
+
"source_appendix_claim_map": dict(self.claim_source_map),
|
|
124
|
+
"source_appendix_trace_generation": self.trace_status,
|
|
125
|
+
"source_appendix_trace_source_count": self.trace_source_count,
|
|
126
|
+
"source_appendix_trace_span_count": self.trace_span_count,
|
|
127
|
+
"source_appendix_trace_warnings": list(self.trace_warnings),
|
|
128
|
+
}
|
|
129
|
+
|
|
130
|
+
|
|
131
|
+
def cited_claim_ids(markdown: str) -> list[str]:
|
|
132
|
+
"""Return [src:<claim_id>] references in first-appearance order."""
|
|
133
|
+
seen: set[str] = set()
|
|
134
|
+
claim_ids: list[str] = []
|
|
135
|
+
for match in _SRC_REF_RE.finditer(markdown):
|
|
136
|
+
claim_id = match.group(1).strip()
|
|
137
|
+
if not claim_id or claim_id in seen:
|
|
138
|
+
continue
|
|
139
|
+
seen.add(claim_id)
|
|
140
|
+
claim_ids.append(claim_id)
|
|
141
|
+
return claim_ids
|
|
142
|
+
|
|
143
|
+
|
|
144
|
+
def build_source_appendix(
|
|
145
|
+
*,
|
|
146
|
+
audited_markdown: str,
|
|
147
|
+
ledger_path: str | Path,
|
|
148
|
+
evidence_span_registry_path: str | Path | None = None,
|
|
149
|
+
workspace: str | Path | None = None,
|
|
150
|
+
) -> SourceAppendixResult:
|
|
151
|
+
"""Build reader-facing source appendix Markdown from cited Claim Ledger entries."""
|
|
152
|
+
claim_ids = cited_claim_ids(audited_markdown)
|
|
153
|
+
warnings: list[str] = []
|
|
154
|
+
ledger = ClaimLedger.import_json(ledger_path)
|
|
155
|
+
records_by_key: dict[str, SourceAppendixRecord] = {}
|
|
156
|
+
claim_source_keys: dict[str, str] = {}
|
|
157
|
+
source_ids_by_key: dict[str, list[str]] = {}
|
|
158
|
+
order: list[str] = []
|
|
159
|
+
resolved_claim_count = 0
|
|
160
|
+
|
|
161
|
+
for claim_id in claim_ids:
|
|
162
|
+
claim = ledger.get_claim(claim_id)
|
|
163
|
+
if claim is None:
|
|
164
|
+
warnings.append("A cited source reference could not be resolved.")
|
|
165
|
+
continue
|
|
166
|
+
resolved_claim_count += 1
|
|
167
|
+
source_record, source_warnings = _record_from_claim(claim)
|
|
168
|
+
warnings.extend(source_warnings)
|
|
169
|
+
key = _source_key(claim, source_record)
|
|
170
|
+
if key not in records_by_key:
|
|
171
|
+
order.append(key)
|
|
172
|
+
records_by_key[key] = source_record
|
|
173
|
+
record = records_by_key[key]
|
|
174
|
+
record.claim_count += 1
|
|
175
|
+
if claim_id not in record.claim_ids:
|
|
176
|
+
record.claim_ids.append(claim_id)
|
|
177
|
+
claim_source_keys[claim_id] = key
|
|
178
|
+
if claim.source_id:
|
|
179
|
+
source_id = claim.source_id.strip()
|
|
180
|
+
if source_id:
|
|
181
|
+
source_ids = source_ids_by_key.setdefault(key, [])
|
|
182
|
+
if source_id not in source_ids:
|
|
183
|
+
source_ids.append(source_id)
|
|
184
|
+
source_claim_ids = record.claim_ids_by_source_id.setdefault(source_id, [])
|
|
185
|
+
if claim_id not in source_claim_ids:
|
|
186
|
+
source_claim_ids.append(claim_id)
|
|
187
|
+
|
|
188
|
+
records = [records_by_key[key] for key in order]
|
|
189
|
+
for idx, record in enumerate(records, start=1):
|
|
190
|
+
record.label = f"S{idx}"
|
|
191
|
+
source_ids = source_ids_by_key.get(order[idx - 1], [])
|
|
192
|
+
record.source_ids = list(source_ids)
|
|
193
|
+
record.source_id = source_ids[0] if source_ids else ""
|
|
194
|
+
citation_labels = {
|
|
195
|
+
claim_id: records_by_key[key].label
|
|
196
|
+
for claim_id, key in claim_source_keys.items()
|
|
197
|
+
if key in records_by_key and records_by_key[key].label
|
|
198
|
+
}
|
|
199
|
+
claim_source_map = {
|
|
200
|
+
claim_id: _claim_source_map_record(records_by_key[key])
|
|
201
|
+
for claim_id, key in claim_source_keys.items()
|
|
202
|
+
if key in records_by_key and records_by_key[key].label
|
|
203
|
+
}
|
|
204
|
+
trace = _build_source_appendix_trace(
|
|
205
|
+
registry_path=evidence_span_registry_path,
|
|
206
|
+
workspace=workspace,
|
|
207
|
+
ledger_path=ledger_path,
|
|
208
|
+
records=records,
|
|
209
|
+
)
|
|
210
|
+
|
|
211
|
+
status = "generated_with_warnings" if warnings else "generated"
|
|
212
|
+
markdown = render_source_appendix(records, warnings=warnings)
|
|
213
|
+
return SourceAppendixResult(
|
|
214
|
+
status=status,
|
|
215
|
+
source_count=len(records),
|
|
216
|
+
cited_claim_count=len(claim_ids),
|
|
217
|
+
resolved_claim_count=resolved_claim_count,
|
|
218
|
+
warnings=warnings,
|
|
219
|
+
markdown=markdown,
|
|
220
|
+
records=records,
|
|
221
|
+
citation_labels=citation_labels,
|
|
222
|
+
claim_source_map=claim_source_map,
|
|
223
|
+
trace_status=trace["status"],
|
|
224
|
+
trace_markdown=trace["markdown"],
|
|
225
|
+
trace_source_count=trace["source_count"],
|
|
226
|
+
trace_span_count=trace["span_count"],
|
|
227
|
+
trace_warnings=trace["warnings"],
|
|
228
|
+
)
|
|
229
|
+
|
|
230
|
+
|
|
231
|
+
def replace_claim_citations_with_labels(
|
|
232
|
+
markdown: str,
|
|
233
|
+
citation_labels: dict[str, str],
|
|
234
|
+
) -> str:
|
|
235
|
+
"""Replace internal claim citations with reader-facing source labels."""
|
|
236
|
+
|
|
237
|
+
def _replace(match: re.Match[str]) -> str:
|
|
238
|
+
label = citation_labels.get(match.group(1).strip())
|
|
239
|
+
return f"[{label}]" if label else ""
|
|
240
|
+
|
|
241
|
+
text = _SRC_REF_RE.sub(_replace, markdown)
|
|
242
|
+
text = re.compile(r"\[src:[^\]]*\]").sub("", text)
|
|
243
|
+
text = re.sub(r"(\[S\d+\])(?:\s+\1)+", r"\1", text)
|
|
244
|
+
text = re.sub(r"\n{3,}", "\n\n", text)
|
|
245
|
+
return text.strip()
|
|
246
|
+
|
|
247
|
+
|
|
248
|
+
def render_source_appendix(
|
|
249
|
+
records: list[SourceAppendixRecord],
|
|
250
|
+
*,
|
|
251
|
+
warnings: list[str] | None = None,
|
|
252
|
+
) -> str:
|
|
253
|
+
lines = [
|
|
254
|
+
"# Source Appendix",
|
|
255
|
+
"",
|
|
256
|
+
"This appendix lists source entries linked from the brief. It is a traceability surface, not semantic proof that every statement is correct.",
|
|
257
|
+
"",
|
|
258
|
+
"## Sources",
|
|
259
|
+
"",
|
|
260
|
+
]
|
|
261
|
+
if not records:
|
|
262
|
+
lines.extend(["No reader-facing sources could be resolved from cited claims.", ""])
|
|
263
|
+
for record in records:
|
|
264
|
+
title = record.title or "Source record"
|
|
265
|
+
lines.append(f"### [{record.label}] {title}")
|
|
266
|
+
lines.append("")
|
|
267
|
+
if record.source_category:
|
|
268
|
+
lines.append(f"- Source category: {_source_category_label(record.source_category)}")
|
|
269
|
+
if record.retrieval_source_type:
|
|
270
|
+
lines.append(
|
|
271
|
+
f"- Retrieval source type: {_retrieval_source_type_label(record.retrieval_source_type)}"
|
|
272
|
+
)
|
|
273
|
+
if record.underlying_evidence_type and record.underlying_evidence_type != "unknown":
|
|
274
|
+
lines.append(
|
|
275
|
+
"- Underlying evidence type: "
|
|
276
|
+
f"{_underlying_evidence_type_label(record.underlying_evidence_type)}"
|
|
277
|
+
)
|
|
278
|
+
if record.publisher:
|
|
279
|
+
lines.append(f"- Publisher/Institution: {record.publisher}")
|
|
280
|
+
if record.published_at:
|
|
281
|
+
lines.append(f"- Published: {record.published_at}")
|
|
282
|
+
if record.retrieved_at:
|
|
283
|
+
lines.append(f"- Retrieved: {record.retrieved_at}")
|
|
284
|
+
if record.url:
|
|
285
|
+
lines.append(f"- URL: <{_markdown_autolink_url(record.url)}>")
|
|
286
|
+
if record.source_type:
|
|
287
|
+
lines.append(f"- Provider type: {record.source_type}")
|
|
288
|
+
lines.append(f"- Used in: {record.claim_count} claim-backed statement{'s' if record.claim_count != 1 else ''}")
|
|
289
|
+
if record.span_count:
|
|
290
|
+
role_summary = ", ".join(_reader_role_label(role) for role in record.span_roles)
|
|
291
|
+
lines.append(
|
|
292
|
+
f"- Evidence trace: {record.span_count} span{'s' if record.span_count != 1 else ''}"
|
|
293
|
+
+ (f"; roles: {role_summary}" if role_summary else "")
|
|
294
|
+
)
|
|
295
|
+
lines.append("")
|
|
296
|
+
if warnings:
|
|
297
|
+
lines.extend([
|
|
298
|
+
"## Appendix Notes",
|
|
299
|
+
"",
|
|
300
|
+
"Some cited source metadata was incomplete or omitted from reader-facing output.",
|
|
301
|
+
"",
|
|
302
|
+
])
|
|
303
|
+
for warning in sorted(dict.fromkeys(warnings)):
|
|
304
|
+
lines.append(f"- {warning}")
|
|
305
|
+
lines.append("")
|
|
306
|
+
return "\n".join(lines).rstrip() + "\n"
|
|
307
|
+
|
|
308
|
+
|
|
309
|
+
def _record_from_claim(claim: Claim) -> tuple[SourceAppendixRecord, list[str]]:
|
|
310
|
+
warnings: list[str] = []
|
|
311
|
+
metadata = claim.metadata or {}
|
|
312
|
+
raw_title = _first_text(
|
|
313
|
+
metadata.get("source_title"),
|
|
314
|
+
metadata.get("title"),
|
|
315
|
+
metadata.get("source_name"),
|
|
316
|
+
)
|
|
317
|
+
raw_publisher = _first_text(metadata.get("publisher"), metadata.get("source_name"))
|
|
318
|
+
title, title_warning = _safe_display_text(raw_title, field_name="source title")
|
|
319
|
+
publisher, publisher_warning = _safe_display_text(raw_publisher, field_name="publisher")
|
|
320
|
+
warnings.extend(item for item in (title_warning, publisher_warning) if item)
|
|
321
|
+
|
|
322
|
+
url, url_warning = _safe_url(claim.source_url or _first_text(metadata.get("url"), metadata.get("source_url")))
|
|
323
|
+
if url_warning:
|
|
324
|
+
warnings.append(url_warning)
|
|
325
|
+
published_at, published_warning = _safe_display_text(
|
|
326
|
+
_first_text(metadata.get("published_at"), metadata.get("publication_date")),
|
|
327
|
+
field_name="published date",
|
|
328
|
+
)
|
|
329
|
+
retrieved_at, retrieved_warning = _safe_display_text(
|
|
330
|
+
_first_text(metadata.get("retrieved_at"), metadata.get("accessed_at")),
|
|
331
|
+
field_name="retrieved date",
|
|
332
|
+
)
|
|
333
|
+
warnings.extend(item for item in (published_warning, retrieved_warning) if item)
|
|
334
|
+
|
|
335
|
+
source_type, type_warning = _safe_display_text(claim.source_type, field_name="source type")
|
|
336
|
+
if type_warning:
|
|
337
|
+
warnings.append(type_warning)
|
|
338
|
+
|
|
339
|
+
source_category, category_warning = _safe_source_category(
|
|
340
|
+
_first_text(metadata.get("source_category"), metadata.get("evidence_category"))
|
|
341
|
+
)
|
|
342
|
+
if category_warning:
|
|
343
|
+
warnings.append(category_warning)
|
|
344
|
+
if not source_category:
|
|
345
|
+
warnings.append("Source metadata missing source category.")
|
|
346
|
+
|
|
347
|
+
raw_retrieval = _first_text(metadata.get("retrieval_source_type"))
|
|
348
|
+
if not raw_retrieval:
|
|
349
|
+
provider_or_storage_type = _first_text(metadata.get("source_type"), claim.source_type)
|
|
350
|
+
if provider_or_storage_type in VALID_RETRIEVAL_SOURCE_TYPES:
|
|
351
|
+
raw_retrieval = provider_or_storage_type
|
|
352
|
+
retrieval_source_type, retrieval_warning = _safe_retrieval_source_type(raw_retrieval)
|
|
353
|
+
if retrieval_warning:
|
|
354
|
+
warnings.append(retrieval_warning)
|
|
355
|
+
|
|
356
|
+
underlying_raw = _first_text(metadata.get("underlying_evidence_type"))
|
|
357
|
+
underlying_evidence_type, underlying_warning = _safe_underlying_evidence_type(underlying_raw)
|
|
358
|
+
if underlying_warning:
|
|
359
|
+
warnings.append(underlying_warning)
|
|
360
|
+
|
|
361
|
+
if not title:
|
|
362
|
+
warnings.append("Source metadata missing source title/name.")
|
|
363
|
+
if not publisher:
|
|
364
|
+
warnings.append("Source metadata missing publisher/institution.")
|
|
365
|
+
|
|
366
|
+
title_for_completeness = title
|
|
367
|
+
if not title:
|
|
368
|
+
title = "Source record"
|
|
369
|
+
|
|
370
|
+
return (
|
|
371
|
+
SourceAppendixRecord(
|
|
372
|
+
label="",
|
|
373
|
+
title=title,
|
|
374
|
+
source_id=claim.source_id.strip() if claim.source_id else "",
|
|
375
|
+
publisher=publisher,
|
|
376
|
+
published_at=published_at,
|
|
377
|
+
retrieved_at=retrieved_at,
|
|
378
|
+
url=url,
|
|
379
|
+
source_type=source_type,
|
|
380
|
+
source_category=source_category,
|
|
381
|
+
retrieval_source_type=retrieval_source_type,
|
|
382
|
+
underlying_evidence_type=underlying_evidence_type,
|
|
383
|
+
metadata_warnings=_metadata_completeness_warnings(
|
|
384
|
+
title=title_for_completeness,
|
|
385
|
+
publisher=publisher,
|
|
386
|
+
source_category=source_category,
|
|
387
|
+
retrieval_source_type=retrieval_source_type,
|
|
388
|
+
underlying_evidence_type=underlying_evidence_type,
|
|
389
|
+
underlying_raw=underlying_raw,
|
|
390
|
+
),
|
|
391
|
+
),
|
|
392
|
+
warnings,
|
|
393
|
+
)
|
|
394
|
+
|
|
395
|
+
|
|
396
|
+
def _claim_source_map_record(record: SourceAppendixRecord) -> dict[str, str]:
|
|
397
|
+
payload = {
|
|
398
|
+
"source_label": record.label,
|
|
399
|
+
"source_url": record.url,
|
|
400
|
+
"evidence_title": record.title,
|
|
401
|
+
"source_title": record.title,
|
|
402
|
+
"source_published_at": record.published_at,
|
|
403
|
+
"retrieved_at": record.retrieved_at,
|
|
404
|
+
"source_type": record.source_type,
|
|
405
|
+
"source_category": record.source_category,
|
|
406
|
+
}
|
|
407
|
+
if record.retrieval_source_type:
|
|
408
|
+
payload["retrieval_source_type"] = record.retrieval_source_type
|
|
409
|
+
if record.underlying_evidence_type:
|
|
410
|
+
payload["underlying_evidence_type"] = record.underlying_evidence_type
|
|
411
|
+
return payload
|
|
412
|
+
|
|
413
|
+
|
|
414
|
+
def _build_source_appendix_trace(
|
|
415
|
+
*,
|
|
416
|
+
registry_path: str | Path | None,
|
|
417
|
+
workspace: str | Path | None,
|
|
418
|
+
ledger_path: str | Path,
|
|
419
|
+
records: list[SourceAppendixRecord],
|
|
420
|
+
) -> dict[str, Any]:
|
|
421
|
+
empty = {
|
|
422
|
+
"status": "not_available",
|
|
423
|
+
"markdown": "",
|
|
424
|
+
"source_count": 0,
|
|
425
|
+
"span_count": 0,
|
|
426
|
+
"warnings": [],
|
|
427
|
+
}
|
|
428
|
+
if registry_path is None:
|
|
429
|
+
return empty
|
|
430
|
+
path = Path(registry_path)
|
|
431
|
+
if not path.exists():
|
|
432
|
+
return empty
|
|
433
|
+
try:
|
|
434
|
+
payload = json.loads(path.read_text(encoding="utf-8"))
|
|
435
|
+
except (OSError, UnicodeDecodeError, json.JSONDecodeError) as exc:
|
|
436
|
+
return _trace_skip(f"Evidence span trace skipped because evidence_span_registry.json is unreadable: {exc}")
|
|
437
|
+
if not isinstance(payload, dict):
|
|
438
|
+
return _trace_skip("Evidence span trace skipped because evidence_span_registry.json is not an object.")
|
|
439
|
+
|
|
440
|
+
violations = EvidenceSpanRegistryContract.validate(payload)
|
|
441
|
+
errors = [violation for violation in violations if violation.severity == "error"]
|
|
442
|
+
if errors:
|
|
443
|
+
return _trace_skip(
|
|
444
|
+
"Evidence span trace skipped because evidence_span_registry.json is invalid: "
|
|
445
|
+
f"evidence_span_registry_schema_error:{errors[0].field}"
|
|
446
|
+
)
|
|
447
|
+
|
|
448
|
+
ws = Path(workspace).expanduser().resolve() if workspace is not None else _workspace_from_ledger_path(ledger_path)
|
|
449
|
+
from multi_agent_brief.orchestrator.runtime_state.evidence_span_registry import (
|
|
450
|
+
EVIDENCE_SPAN_REGISTRY_VALIDATION_PREFIX,
|
|
451
|
+
validate_evidence_span_registry_against_source_pack,
|
|
452
|
+
)
|
|
453
|
+
|
|
454
|
+
reason = validate_evidence_span_registry_against_source_pack(
|
|
455
|
+
registry_payload=payload,
|
|
456
|
+
workspace=ws,
|
|
457
|
+
)
|
|
458
|
+
if reason:
|
|
459
|
+
return _trace_skip(
|
|
460
|
+
"Evidence span trace skipped because evidence_span_registry.json does not match source bytes: "
|
|
461
|
+
f"{EVIDENCE_SPAN_REGISTRY_VALIDATION_PREFIX}:{reason}"
|
|
462
|
+
)
|
|
463
|
+
|
|
464
|
+
records_by_source_id: dict[str, SourceAppendixRecord] = {}
|
|
465
|
+
trace_spans_by_label: dict[str, int] = {}
|
|
466
|
+
trace_roles_by_label: dict[str, set[str]] = {}
|
|
467
|
+
trace_sources: list[dict[str, Any]] = []
|
|
468
|
+
for record in records:
|
|
469
|
+
source_ids = record.source_ids or ([record.source_id] if record.source_id else [])
|
|
470
|
+
for source_id in source_ids:
|
|
471
|
+
normalized = str(source_id or "").strip()
|
|
472
|
+
if normalized:
|
|
473
|
+
records_by_source_id[normalized] = record
|
|
474
|
+
for source in sorted(
|
|
475
|
+
(item for item in payload.get("sources", []) if isinstance(item, dict)),
|
|
476
|
+
key=lambda item: str(item.get("source_id") or ""),
|
|
477
|
+
):
|
|
478
|
+
source_id = str(source.get("source_id") or "").strip()
|
|
479
|
+
record = records_by_source_id.get(source_id)
|
|
480
|
+
if record is None:
|
|
481
|
+
continue
|
|
482
|
+
spans = [
|
|
483
|
+
span
|
|
484
|
+
for span in source.get("spans", [])
|
|
485
|
+
if isinstance(span, dict)
|
|
486
|
+
]
|
|
487
|
+
if not spans:
|
|
488
|
+
continue
|
|
489
|
+
source_path = str(source.get("source_path") or "")
|
|
490
|
+
snapshot, snapshot_warning = _source_snapshot(ws=ws, source_path=source_path)
|
|
491
|
+
metadata_warnings = list(record.metadata_warnings)
|
|
492
|
+
if snapshot_warning:
|
|
493
|
+
metadata_warnings.append(snapshot_warning)
|
|
494
|
+
roles = sorted({
|
|
495
|
+
str(span.get("span_role") or "").strip()
|
|
496
|
+
for span in spans
|
|
497
|
+
if str(span.get("span_role") or "").strip()
|
|
498
|
+
})
|
|
499
|
+
trace_spans_by_label[record.label] = trace_spans_by_label.get(record.label, 0) + len(spans)
|
|
500
|
+
trace_roles_by_label.setdefault(record.label, set()).update(roles)
|
|
501
|
+
trace_sources.append({
|
|
502
|
+
"label": record.label,
|
|
503
|
+
"title": record.title,
|
|
504
|
+
"source_id": source_id,
|
|
505
|
+
"source_path": source_path,
|
|
506
|
+
"claim_ids": list(record.claim_ids_by_source_id.get(source_id) or record.claim_ids),
|
|
507
|
+
"source_sha256": snapshot.get("sha256", ""),
|
|
508
|
+
"source_size_bytes": snapshot.get("size_bytes", 0),
|
|
509
|
+
"metadata_warnings": metadata_warnings,
|
|
510
|
+
"spans": sorted(spans, key=lambda item: str(item.get("span_id") or "")),
|
|
511
|
+
})
|
|
512
|
+
|
|
513
|
+
if not trace_sources:
|
|
514
|
+
return _trace_skip("Evidence span trace skipped because no registry sources matched cited claims.")
|
|
515
|
+
|
|
516
|
+
for record in records:
|
|
517
|
+
if not record.label:
|
|
518
|
+
continue
|
|
519
|
+
record.span_count = trace_spans_by_label.get(record.label, 0)
|
|
520
|
+
record.span_roles = sorted(trace_roles_by_label.get(record.label, set()))
|
|
521
|
+
|
|
522
|
+
markdown = render_source_appendix_trace(trace_sources)
|
|
523
|
+
return {
|
|
524
|
+
"status": "generated",
|
|
525
|
+
"markdown": markdown,
|
|
526
|
+
"source_count": len(trace_sources),
|
|
527
|
+
"span_count": sum(len(source["spans"]) for source in trace_sources),
|
|
528
|
+
"warnings": [],
|
|
529
|
+
}
|
|
530
|
+
|
|
531
|
+
|
|
532
|
+
def _trace_skip(message: str) -> dict[str, Any]:
|
|
533
|
+
return {
|
|
534
|
+
"status": "skipped",
|
|
535
|
+
"markdown": "",
|
|
536
|
+
"source_count": 0,
|
|
537
|
+
"span_count": 0,
|
|
538
|
+
"warnings": [message],
|
|
539
|
+
}
|
|
540
|
+
|
|
541
|
+
|
|
542
|
+
def render_source_appendix_trace(trace_sources: list[dict[str, Any]]) -> str:
|
|
543
|
+
lines = [
|
|
544
|
+
"# Evidence Span Trace Audit Copy",
|
|
545
|
+
"",
|
|
546
|
+
"This audit copy lists machine-checkable Evidence Span Registry entries used by the source appendix. It is a traceability surface only, not semantic proof or support sufficiency.",
|
|
547
|
+
"",
|
|
548
|
+
]
|
|
549
|
+
for source in trace_sources:
|
|
550
|
+
title = source.get("title") or "Local workspace source"
|
|
551
|
+
lines.extend([
|
|
552
|
+
f"## [{source.get('label')}] {title}",
|
|
553
|
+
"",
|
|
554
|
+
f"- Source ID: `{source.get('source_id')}`",
|
|
555
|
+
f"- Source path: `{source.get('source_path')}`",
|
|
556
|
+
"- Claim IDs: " + _inline_code_list(source.get("claim_ids") or []),
|
|
557
|
+
f"- Source SHA-256: `{source.get('source_sha256') or ''}`",
|
|
558
|
+
f"- Source size bytes: {source.get('source_size_bytes') or 0}",
|
|
559
|
+
f"- Span count: {len(source.get('spans') or [])}",
|
|
560
|
+
"",
|
|
561
|
+
])
|
|
562
|
+
warnings = [str(item).strip() for item in source.get("metadata_warnings") or [] if str(item).strip()]
|
|
563
|
+
if warnings:
|
|
564
|
+
lines.extend(["Metadata warnings:", ""])
|
|
565
|
+
for warning in sorted(dict.fromkeys(warnings)):
|
|
566
|
+
lines.append(f"- {warning}")
|
|
567
|
+
lines.append("")
|
|
568
|
+
for span in source.get("spans") or []:
|
|
569
|
+
span_id = str(span.get("span_id") or "<unknown_span>").strip()
|
|
570
|
+
lines.extend([
|
|
571
|
+
f"### {span_id}",
|
|
572
|
+
"",
|
|
573
|
+
f"- Role: {_reader_role_label(str(span.get('span_role') or ''))}",
|
|
574
|
+
f"- Raw excerpt hash: `{span.get('hash') or ''}`",
|
|
575
|
+
])
|
|
576
|
+
if isinstance(span.get("char_start"), int) and isinstance(span.get("char_end"), int):
|
|
577
|
+
lines.append(f"- Offsets: {span['char_start']}..{span['char_end']}")
|
|
578
|
+
lines.extend(["", "Raw excerpt:", ""])
|
|
579
|
+
lines.extend(_blockquote(_cap_excerpt(str(span.get("raw_excerpt") or ""))))
|
|
580
|
+
lines.append("")
|
|
581
|
+
return "\n".join(lines).rstrip() + "\n"
|
|
582
|
+
|
|
583
|
+
|
|
584
|
+
def _workspace_from_ledger_path(ledger_path: str | Path) -> Path:
|
|
585
|
+
path = Path(ledger_path).expanduser().resolve()
|
|
586
|
+
try:
|
|
587
|
+
return path.parents[2]
|
|
588
|
+
except IndexError:
|
|
589
|
+
return path.parent
|
|
590
|
+
|
|
591
|
+
|
|
592
|
+
def _reader_role_label(role: str) -> str:
|
|
593
|
+
return role.replace("_", " ").strip()
|
|
594
|
+
|
|
595
|
+
|
|
596
|
+
def _source_category_label(category: str) -> str:
|
|
597
|
+
return _SOURCE_CATEGORY_LABELS.get(category, category.replace("_", " ").strip().title())
|
|
598
|
+
|
|
599
|
+
|
|
600
|
+
def _retrieval_source_type_label(value: str) -> str:
|
|
601
|
+
return _RETRIEVAL_SOURCE_TYPE_LABELS.get(value, value.replace("_", " ").strip().title())
|
|
602
|
+
|
|
603
|
+
|
|
604
|
+
def _underlying_evidence_type_label(value: str) -> str:
|
|
605
|
+
return _UNDERLYING_EVIDENCE_TYPE_LABELS.get(value, value.replace("_", " ").strip().title())
|
|
606
|
+
|
|
607
|
+
|
|
608
|
+
def _safe_retrieval_source_type(value: str) -> tuple[str, str]:
|
|
609
|
+
raw = str(value or "").strip()
|
|
610
|
+
if not raw:
|
|
611
|
+
return "", ""
|
|
612
|
+
if raw not in VALID_RETRIEVAL_SOURCE_TYPES:
|
|
613
|
+
return "", "Omitted retrieval source type because it was not recognized."
|
|
614
|
+
return raw, ""
|
|
615
|
+
|
|
616
|
+
|
|
617
|
+
def _safe_underlying_evidence_type(value: str) -> tuple[str, str]:
|
|
618
|
+
raw = str(value or "").strip()
|
|
619
|
+
if not raw:
|
|
620
|
+
return "", ""
|
|
621
|
+
if raw not in VALID_UNDERLYING_EVIDENCE_TYPES:
|
|
622
|
+
return "", "Omitted underlying evidence type because it was not recognized."
|
|
623
|
+
return raw, ""
|
|
624
|
+
|
|
625
|
+
|
|
626
|
+
def _metadata_completeness_warnings(
|
|
627
|
+
*,
|
|
628
|
+
title: str,
|
|
629
|
+
publisher: str,
|
|
630
|
+
source_category: str,
|
|
631
|
+
retrieval_source_type: str,
|
|
632
|
+
underlying_evidence_type: str,
|
|
633
|
+
underlying_raw: str,
|
|
634
|
+
) -> list[str]:
|
|
635
|
+
warnings: list[str] = []
|
|
636
|
+
if not title:
|
|
637
|
+
warnings.append("Source metadata missing source title/name.")
|
|
638
|
+
if not publisher:
|
|
639
|
+
warnings.append("Source metadata missing publisher/institution.")
|
|
640
|
+
if not source_category:
|
|
641
|
+
warnings.append("Source metadata missing source category.")
|
|
642
|
+
if not retrieval_source_type:
|
|
643
|
+
warnings.append("Source metadata missing retrieval source type.")
|
|
644
|
+
if not underlying_raw:
|
|
645
|
+
warnings.append("Source metadata missing underlying evidence type.")
|
|
646
|
+
elif underlying_evidence_type == "unknown":
|
|
647
|
+
warnings.append("Source metadata underlying evidence type is unknown.")
|
|
648
|
+
return warnings
|
|
649
|
+
|
|
650
|
+
|
|
651
|
+
def _source_snapshot(*, ws: Path, source_path: str) -> tuple[dict[str, Any], str]:
|
|
652
|
+
if not source_path:
|
|
653
|
+
return {}, "Source snapshot unavailable because source path is missing."
|
|
654
|
+
try:
|
|
655
|
+
path = (ws / source_path).resolve()
|
|
656
|
+
root = (ws / "input" / "sources").resolve()
|
|
657
|
+
path.relative_to(root)
|
|
658
|
+
data = path.read_bytes()
|
|
659
|
+
except (OSError, ValueError):
|
|
660
|
+
return {}, "Source snapshot unavailable because source bytes could not be read."
|
|
661
|
+
return {
|
|
662
|
+
"sha256": hashlib.sha256(data).hexdigest(),
|
|
663
|
+
"size_bytes": len(data),
|
|
664
|
+
}, ""
|
|
665
|
+
|
|
666
|
+
|
|
667
|
+
def _inline_code_list(values: list[Any]) -> str:
|
|
668
|
+
items = [f"`{str(value).strip()}`" for value in values if str(value).strip()]
|
|
669
|
+
return ", ".join(items) if items else "`<none>`"
|
|
670
|
+
|
|
671
|
+
|
|
672
|
+
def _cap_excerpt(text: str) -> str:
|
|
673
|
+
if len(text) <= _TRACE_EXCERPT_LIMIT:
|
|
674
|
+
return text
|
|
675
|
+
return text[:_TRACE_EXCERPT_LIMIT].rstrip() + "... [truncated]"
|
|
676
|
+
|
|
677
|
+
|
|
678
|
+
def _blockquote(text: str) -> list[str]:
|
|
679
|
+
if not text:
|
|
680
|
+
return [">"]
|
|
681
|
+
return ["> " + line for line in text.splitlines()]
|
|
682
|
+
|
|
683
|
+
|
|
684
|
+
def _source_key(claim: Claim, record: SourceAppendixRecord) -> str:
|
|
685
|
+
if record.url:
|
|
686
|
+
return f"url:{_normalize(record.url)}"
|
|
687
|
+
if claim.source_id:
|
|
688
|
+
return f"source_id:{_normalize(claim.source_id)}"
|
|
689
|
+
return f"title:{_normalize(record.title)}|publisher:{_normalize(record.publisher)}"
|
|
690
|
+
|
|
691
|
+
|
|
692
|
+
def _first_text(*values: Any) -> str:
|
|
693
|
+
for value in values:
|
|
694
|
+
if isinstance(value, str) and value.strip():
|
|
695
|
+
return value.strip()
|
|
696
|
+
return ""
|
|
697
|
+
|
|
698
|
+
|
|
699
|
+
def _safe_display_text(value: str, *, field_name: str) -> tuple[str, str]:
|
|
700
|
+
value = " ".join(str(value or "").split())
|
|
701
|
+
if not value:
|
|
702
|
+
return "", ""
|
|
703
|
+
if _contains_private_path(value):
|
|
704
|
+
return "", f"Omitted {field_name} because it looked like a local path."
|
|
705
|
+
if value.lower().startswith("file://"):
|
|
706
|
+
return "", f"Omitted {field_name} because it used a file:// reference."
|
|
707
|
+
if _contains_internal_id(value):
|
|
708
|
+
return "", f"Omitted {field_name} because it looked like an internal ID."
|
|
709
|
+
return value, ""
|
|
710
|
+
|
|
711
|
+
|
|
712
|
+
def _safe_url(value: str) -> tuple[str, str]:
|
|
713
|
+
raw = str(value or "").strip()
|
|
714
|
+
if not raw:
|
|
715
|
+
return "", ""
|
|
716
|
+
if _contains_private_path(raw) or raw.lower().startswith("file://"):
|
|
717
|
+
return "", "Omitted source URL because it was local or file-based."
|
|
718
|
+
parsed = urlparse(raw)
|
|
719
|
+
if parsed.scheme not in {"http", "https"} or not parsed.netloc:
|
|
720
|
+
return "", "Omitted source URL because it was not an HTTP(S) URL."
|
|
721
|
+
return raw, ""
|
|
722
|
+
|
|
723
|
+
|
|
724
|
+
def _markdown_autolink_url(value: str) -> str:
|
|
725
|
+
return str(value).replace("<", "%3C").replace(">", "%3E").replace(" ", "%20")
|
|
726
|
+
|
|
727
|
+
|
|
728
|
+
def _safe_source_category(value: str) -> tuple[str, str]:
|
|
729
|
+
raw = str(value or "").strip()
|
|
730
|
+
if not raw:
|
|
731
|
+
return "", ""
|
|
732
|
+
if raw not in VALID_SOURCE_CATEGORIES:
|
|
733
|
+
return "", "Omitted source category because it was not recognized."
|
|
734
|
+
return raw, ""
|
|
735
|
+
|
|
736
|
+
|
|
737
|
+
def _contains_private_path(value: str) -> bool:
|
|
738
|
+
if any(marker in value for marker in _LOCAL_PATH_MARKERS):
|
|
739
|
+
return True
|
|
740
|
+
if _WINDOWS_USER_RE.search(value):
|
|
741
|
+
return True
|
|
742
|
+
try:
|
|
743
|
+
return PureWindowsPath(value).is_absolute()
|
|
744
|
+
except ValueError:
|
|
745
|
+
return False
|
|
746
|
+
|
|
747
|
+
|
|
748
|
+
def _contains_internal_id(value: str) -> bool:
|
|
749
|
+
return bool(_INTERNAL_ID_RE.search(value))
|
|
750
|
+
|
|
751
|
+
|
|
752
|
+
def _normalize(value: str) -> str:
|
|
753
|
+
return " ".join(str(value or "").strip().lower().split())
|