Sprint 0+1: Package restructure, source tiers, verdicts, multi-variant editions

- New oracle/ package (11 modules) with unified CLI (python -m oracle)
- Source tiers: Tier 1 (arxiv/github/hf), Tier 2 (rss/hn), Tier 3 (reddit)
- Composite verdicts: PUBLISH/WATCH/ARCHIVE/DROP based on signal score + age
- Content-hash dedup: SHA-256[:16] normalized, atomic at insert time
- Multi-variant editions: 4 YAML configs (default/research/devops/brief)
- Variant engine: filter → rank → render (HTML + JSON, themed)
- Per-adapter timeout (10s) + threading fallback
- Consolidated 12 root scripts → thin wrappers + oracle/ package
- Archived stale scripts (_engagement, _live_compare, reddit_proof)
- Updated .gitignore, README.md, schema.sql

adapters/_store.py

@@ -89,8 +89,9 @@ def true_first_seen(raw_meta, source, now_str):
 UPSERT_SQL = """
 INSERT INTO entries
   (source, source_id, url, title, extracted_text, summary,
-   category_tags, signal_score, raw_metadata, first_seen, last_updated)
-VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+   category_tags, signal_score, raw_metadata, first_seen, last_updated,
+   content_hash, source_tier, verdict)
+VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
 ON CONFLICT(source, source_id) DO UPDATE SET
   source      = excluded.source,
   source_id   = excluded.source_id,
@@ -101,6 +102,9 @@ ON CONFLICT(source, source_id) DO UPDATE SET
   signal_score   = excluded.signal_score,
   raw_metadata    = excluded.raw_metadata,
   last_updated    = excluded.last_updated,
+  content_hash    = excluded.content_hash,
+  source_tier     = excluded.source_tier,
+  verdict         = excluded.verdict,
   first_seen  = COALESCE((SELECT first_seen FROM entries WHERE source = excluded.source AND source_id = excluded.source_id), excluded.first_seen)
 """
 
@@ -111,16 +115,38 @@ def upsert_entries(conn, entries):
     `entries` is the list of dicts as built by each adapter; each dict must
     already have first_seen set to the TRUE publish date (via true_first_seen)
     and last_updated to the harvest time.
+
+    Now also sets content_hash and source_tier at insertion time.
     Returns count of rows written.
     """
+    from oracle.dedup import content_hash, get_source_tier, compute_verdict, age_hours
+
     cur = conn.cursor()
     written = 0
     for e in entries:
+        # Compute content hash
+        title = e.get("title", "")
+        url = e.get("url", "")
+        body = e.get("extracted_text", "")[:500]
+        h = content_hash(title, url, body)
+
+        # Get source tier
+        source = e.get("source", "")
+        tier_info = get_source_tier(source)
+        tier = tier_info["tier"]
+
+        # Compute verdict
+        first_seen = e.get("first_seen", "")
+        score = float(e.get("signal_score") or 0)
+        age = age_hours(first_seen)
+        verdict = compute_verdict(score, age)
+
         cur.execute(UPSERT_SQL, (
             e["source"], e["source_id"], e["url"], e["title"],
             e.get("extracted_text"), e.get("summary"),
             e.get("category_tags"), e.get("signal_score"),
             e.get("raw_metadata"), e["first_seen"], e["last_updated"],
+            h, tier, verdict,
         ))
         written += 1
     conn.commit()