4.0 KiB
4.0 KiB
Changed Files Manifest
| File | Change Type | Why Changed | Related Guidance | Need Human Review |
|---|---|---|---|---|
docs/CONTENT_STABILIZATION_PROTOCOL.md |
added | Define content stabilization gates | v0.3 Task 1 | yes |
sources/evidence_coverage.matrix.json |
added | Field-level evidence coverage matrix | v0.3 Task 2 | yes |
reports/evidence_coverage_report_v0.2.md |
added | Human-readable evidence coverage summary | v0.3 Task 2 | yes |
sources/source_excerpts.json |
modified | Add quote_status/source_location and QPI pathology excerpts | v0.3 Task 3 | yes |
schemas/source_excerpt.schema.json |
modified | Formalize quote_status/source_location and excerpt_type enum | v0.3 Task 3 | yes |
schemas/source_article.schema.json |
modified | Formalize source_type/source_status enum | v0.3 Task 3 | yes |
models/qpi.model.json |
modified | Strengthen QPI output contract and source evidence count | v0.3 Task 4 | yes |
cards/qpi.md |
modified | Mirror QPI output contract and evidence references | v0.3 Task 4 | yes |
models/intellectual_archaeology.model.json |
modified | Strengthen IA stop gate and structured output contract | v0.3 Task 5 | yes |
cards/intellectual_archaeology.md |
modified | Mirror IA stop gate and output contract | v0.3 Task 5 | yes |
schemas/regression_case.schema.json |
modified | Add v0.2 evaluable fields and new case types | v0.3 Task 7 | yes |
tests/qpi.regression.json |
added | Per-model QPI regression source of truth, 17 cases | v0.3 Task 6 | yes |
tests/intellectual_archaeology.regression.json |
added | Per-model IA regression source of truth, 17 cases | v0.3 Task 6 | yes |
tests/regression_cases.json |
modified | Aggregate regression cases for validation | v0.3 Task 6 | yes |
tests/regression_index.json |
added | Audit index for split regression files | GPT audit package | yes |
selector/selector_rules.json |
modified | Selector v0.2 rule config | v0.3 Task 8 | yes |
selector/selector_examples.json |
modified | Add no-call and QPI-before-IA examples | v0.3 Task 8 | yes |
scripts/run_selector_demo.py |
modified | Emit selected/rejected models, reasons, penalties, no_call | v0.3 Task 8 | no |
scripts/run_selector_regression.py |
added | Validate no-call/selector gate/pipeline cases | v0.3 Task 8 | no |
scripts/check_model_card_sync.py |
added | Check hard-field JSON/Markdown sync | v0.3 Task 9 | no |
scripts/validate_model_library.py |
modified | Validate quote status, enums, 15-case coverage, v0.2 fields | v0.3 Tasks 3/7 | no |
models/model_index.json |
regenerated | Sync evidence and regression counts | v0.2 index strategy | yes |
cards/card_index.md |
regenerated | Human-readable projection of model index | v0.2 index strategy | yes |
reports/content_review_report_v0.2.md |
added | Content stabilization review summary | v0.3 Task 10 | yes |
reports/model_review_status.json |
added | Structured review status | v0.3 Task 10 | yes |
reports/next_action_register_v0.2.md |
added | Next review/action items | v0.3 Task 10 | yes |
reports/selector_regression_report_v0.2.md |
added/generated | Selector regression PASS report | v0.3 Task 8 | yes |
reports/model_card_sync_report_v0.2.md |
added/generated | JSON/Markdown sync PASS report | v0.3 Task 9 | yes |
reports/validation_report.md |
generated | Validator result | Validation | no |
reports/index_rebuild_report.md |
generated | Index check result | v0.2 index strategy | no |
reports/ChatGPT交接文档_模型库MVP_第二轮内容稳定化_2026-06-16.md |
added | Audit handoff package | GPT audit request | yes |
reports/command_log.md |
added | Command results for audit | GPT audit request | no |
reports/open_questions_for_CCRA.md |
added | Concrete review questions | GPT audit request | yes |
reports/schema_change_notes.md |
added | Schema diff explanation | GPT audit request | yes |
repo_tree.txt |
added | 2-level repo structure for audit | GPT audit request | no |