Add dataset card, freeze manifest, and pipeline manifests (snapshot 2026-08-27)

.gitattributes

@@ -60,3 +60,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
 extracted/entities.json filter=lfs diff=lfs merge=lfs -text
 extracted/embeddings.csv filter=lfs diff=lfs merge=lfs -text
+catalogs/delineation-manifest.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -14,7 +14,7 @@ tags:
   - epstein
   - efta
 size_categories:
-  - 1K<n<10K
+  - 10K<n<100K
 ---
 
 # Epstein Files Corpus — OCR'd & Cleaned
@@ -26,15 +26,15 @@ USVI v. JPMorgan, and others), and FOIA productions (FBI, BOP, CBP, Florida).
 All content is public-record material released by the U.S. Department of
 Justice, federal/state courts, and FOIA respondents.
 
-This is a **living dataset**: this snapshot publishes 9,269 documents;
+This is a **living dataset**: this snapshot publishes 32,172 documents — including the full House Oversight estate releases of Oct 17 (8,718 pages) and Nov 12 (22,903 pages);
 roughly 33,000 source PDFs are downloaded and mid-OCR at snapshot time,
 with over 1.3M catalog entries identified across 73 source datasets.
 
 ## Repository layout
 
 ```
-cleaned/<dataset>/<document_id>.md     # 9,269 cleaned OCR markdown documents
-extracted/summaries/<dataset>/<id>.txt # 8,815 extractive summaries
+cleaned/<dataset>/<document_id>.md     # 32,172 cleaned OCR markdown documents
+extracted/summaries/<dataset>/<id>.txt # 30,888 extractive summaries
 extracted/embeddings.csv               # chunk embeddings (document_id, chunk, vector)
 extracted/entities.json                # extracted entity index
 catalogs/freeze-manifest-2026-08-27.json  # acquisition freeze: per-dataset locator counts

catalogs/clean-manifest.json

@@ -3,15 +3,15 @@
     "description": "Cleaned documents with OCR artifacts removed, headers normalized, and formatting fixed.",
     "delineated_dir": "/home/robbd/my/apps/epstein-index/data/delineated",
     "output_dir": "/home/robbd/my/apps/epstein-index/data/cleaned",
-    "total_processed": 9269
+    "total_processed": 32172
   },
   "stats": {
     "skipped": 0,
-    "disclaimers_removed": 195,
-    "redactions_normalized": 14,
-    "stamps_removed": 522,
-    "bates_removed": 2291,
-    "headers_normalized": 8596,
-    "titles_added": 9190
+    "disclaimers_removed": 1962,
+    "redactions_normalized": 22,
+    "stamps_removed": 1293,
+    "bates_removed": 2741,
+    "headers_normalized": 29934,
+    "titles_added": 32091
   }
 }