api.py: enable PRAGMA foreign_keys=ON in _connect helper; clean up 2 message orphans

The messages table declares FOREIGN KEY (conversation_id) REFERENCES
conversations(id), but PRAGMA foreign_keys was never enabled — SQLite
defaults it to OFF per connection, and _connect() did not set it. Two
orphan rows existed in messages (conversation_id='test123' pointing at
a never-existing conversation; both rows from one ~11-second test event
on 2026-04-26).

Audit before changing the PRAGMA:
- All FOREIGN KEY declarations across both DBs (conversations.db,
  sessions.db) accounted for via PRAGMA foreign_key_list on each
  table. Only one FK exists: messages.conversation_id ->
  conversations.id, ON DELETE NO ACTION.
- All tables enumerated via sqlite_master. Two tables in
  conversations.db (conversations, messages); one in sessions.db
  (sessions). No surprises.
- PRAGMA foreign_key_check confirmed exactly the 2 known orphans and
  zero violations elsewhere.

Both delete paths in api.py (delete_conversation at :471, and
clear_all_conversations at :986) already delete from messages BEFORE
conversations, so cascade behavior was correct in code. The orphan
state was caused by a direct INSERT against a non-existent
conversation_id at chat-test time, which an unenforced FK silently
accepted. Turning the PRAGMA on prevents this class of bug at insert
time, not delete time — no delete-path code changes were needed.

Order of operations followed the constraint that orphan cleanup must
precede PRAGMA-on (SQLite would not retroactively delete orphans, but
foreign_key_check would surface them confusingly on any future
operation that touched the messages table):
1. DELETE FROM messages WHERE conversation_id NOT IN (SELECT id FROM
   conversations) — removed the 2 known orphans.
2. Added PRAGMA foreign_keys=ON to _connect() so every connection
   from _connect_conversations() and _connect_sessions() gets FK
   enforcement (SQLite requires per-connection setting).
3. Restarted aaronai.service.

Verification:
- Smoke: GET /api/conversations and /api/conversations/{id}/messages
  both return 200 with expected payloads against the live api.
- E2E single-delete: synthetic conversation + 2 messages inserted via
  the api's _connect helper (FK on); DELETE /api/conversations/{id}
  via the live endpoint removed both rows from both tables.
- Clear-all e2e: skipped on live DB (destructive) — code shape is
  structurally identical to single-delete, no FK-relevant logic
  difference.
- Load-bearing negative test: INSERT into messages with a
  non-existent conversation_id via _connect_conversations() raised
  sqlite3.IntegrityError("FOREIGN KEY constraint failed"). This is
  what proves the PRAGMA actually took effect, not just that we set
  it.

Final counts: 7 conversations, 290 messages (down from 292 by the 2
orphans cleaned up).

Note: an explicit BEGIN/COMMIT around the two-execute delete paths
was considered and skipped. SQLite's implicit-transactional default
already gives the atomicity needed; explicit transactions would be
clarity-only and belong in a separate commit.

scripts/api.py

@@ -38,6 +38,19 @@ load_dotenv(Path.home() / "aaronai" / ".env")
 
 MEMORY_PATH = Path.home() / "aaronai" / "memory.md"
 CONVERSATIONS_DB = str(Path.home() / "aaronai" / "conversations.db")
+
+def _connect(path):
+    conn = sqlite3.connect(path, timeout=5.0)
+    conn.execute("PRAGMA synchronous=NORMAL")
+    conn.execute("PRAGMA foreign_keys=ON")
+    return conn
+
+def _connect_conversations():
+    return _connect(CONVERSATIONS_DB)
+
+def _connect_sessions():
+    return _connect(SESSIONS_DB)
+
 SETTINGS_PATH = Path.home() / "aaronai" / "settings.json"
 WATCHER_LOG = str(Path.home() / "aaronai" / "watcher.log")
 WATCHER_STATE = str(Path.home() / "aaronai" / "watcher_state.json")
@@ -73,7 +86,7 @@ WHISPER_PROMPT = (
 whisper_model = None
 if HAS_WHISPER:
     try:
-        whisper_model = WhisperModel("large-v3", device="cpu", compute_type="int8", cpu_threads=8)
+        whisper_model = WhisperModel("distil-large-v3", device="cpu", compute_type="int8", cpu_threads=4)
         print("Whisper model loaded")
     except Exception as e:
         print(f"Whisper not available: {e}")
@@ -122,8 +135,9 @@ SESSION_PASSWORD = os.getenv("AARON_AI_PASSWORD", "changeme")
 SESSIONS_DB = str(Path.home() / "aaronai" / "sessions.db")
 
 def _init_sessions():
-    conn = sqlite3.connect(SESSIONS_DB)
+    conn = _connect_sessions()
     conn.execute("CREATE TABLE IF NOT EXISTS sessions (token TEXT PRIMARY KEY, created_at TEXT)")
+    conn.execute("PRAGMA journal_mode=WAL")
     conn.commit()
     conn.close()
 
@@ -136,19 +150,19 @@ def hash_password(password: str) -> str:
     return hashlib.sha256(password.encode()).hexdigest()
 
 def save_session(token: str):
-    conn = sqlite3.connect(SESSIONS_DB)
+    conn = _connect_sessions()
     conn.execute("INSERT OR REPLACE INTO sessions VALUES (?, ?)", (token, datetime.now().isoformat()))
     conn.commit()
     conn.close()
 
 def delete_session(token: str):
-    conn = sqlite3.connect(SESSIONS_DB)
+    conn = _connect_sessions()
     conn.execute("DELETE FROM sessions WHERE token = ?", (token,))
     conn.commit()
     conn.close()
 
 def session_exists(token: str) -> bool:
-    conn = sqlite3.connect(SESSIONS_DB)
+    conn = _connect_sessions()
     row = conn.execute("SELECT 1 FROM sessions WHERE token = ?", (token,)).fetchone()
     conn.close()
     return row is not None
@@ -163,7 +177,7 @@ def require_auth(request: Request):
     return token
 
 def init_conversations_db():
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute('''CREATE TABLE IF NOT EXISTS conversations (
         id TEXT PRIMARY KEY,
@@ -182,6 +196,8 @@ def init_conversations_db():
         timestamp TEXT NOT NULL,
         FOREIGN KEY (conversation_id) REFERENCES conversations(id)
     )''')
+    c.execute("PRAGMA journal_mode=WAL")
+    c.execute("CREATE INDEX IF NOT EXISTS idx_messages_conv_ts ON messages(conversation_id, timestamp DESC)")
     conn.commit()
     conn.close()
 
@@ -250,7 +266,7 @@ def retrieve_context(query, n_results=8):
     return context_pieces, sources
 
 def get_conversation_history(conversation_id, limit=20):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute('''SELECT role, content FROM messages
                  WHERE conversation_id = ?
@@ -260,7 +276,7 @@ def get_conversation_history(conversation_id, limit=20):
     return [{"role": r[0], "content": r[1]} for r in reversed(rows)]
 
 def save_message(conversation_id, role, content, sources=None):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     msg_id = hashlib.md5(f"{conversation_id}{role}{datetime.now().isoformat()}".encode()).hexdigest()
     timestamp = datetime.now().isoformat()
@@ -274,7 +290,7 @@ def save_message(conversation_id, role, content, sources=None):
     conn.close()
 
 def create_conversation(title="New conversation"):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     conv_id = hashlib.md5(f"{datetime.now().isoformat()}".encode()).hexdigest()[:16]
     now = datetime.now().isoformat()
@@ -409,7 +425,7 @@ async def update_settings(request: Request, auth: str = Depends(require_auth)):
 
 @app.get("/api/conversations")
 async def list_conversations(auth: str = Depends(require_auth)):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute('''SELECT id, title, created_at, updated_at, message_count
                  FROM conversations ORDER BY updated_at DESC LIMIT 100''')
@@ -429,7 +445,7 @@ async def new_conversation(request: Request, auth: str = Depends(require_auth)):
 
 @app.get("/api/conversations/{conv_id}/messages")
 async def get_messages(conv_id: str, auth: str = Depends(require_auth)):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute('''SELECT role, content, sources, timestamp FROM messages
                  WHERE conversation_id = ? ORDER BY timestamp ASC''', (conv_id,))
@@ -446,7 +462,7 @@ async def rename_conversation(conv_id: str, request: Request, auth: str = Depend
     title = data.get("title", "")
     if not title:
         return JSONResponse({"error": "Title required"}, status_code=400)
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute("UPDATE conversations SET title = ? WHERE id = ?", (title, conv_id))
     conn.commit()
@@ -455,7 +471,7 @@ async def rename_conversation(conv_id: str, request: Request, auth: str = Depend
 
 @app.delete("/api/conversations/{conv_id}")
 async def delete_conversation(conv_id: str, auth: str = Depends(require_auth)):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute("DELETE FROM messages WHERE conversation_id = ?", (conv_id,))
     c.execute("DELETE FROM conversations WHERE id = ?", (conv_id,))
@@ -500,14 +516,14 @@ async def chat_endpoint(request: Request, auth: str = Depends(require_auth)):
     save_message(conversation_id, "user", user_message)
 
     # Auto-title conversation from first message
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute("SELECT message_count, title FROM conversations WHERE id = ?", (conversation_id,))
     row = c.fetchone()
     conn.close()
     if row and row[0] <= 1 and row[1] == "New conversation":
         auto_title = user_message[:60] + ("..." if len(user_message) > 60 else "")
-        conn = sqlite3.connect(CONVERSATIONS_DB)
+        conn = _connect_conversations()
         c = conn.cursor()
         c.execute("UPDATE conversations SET title = ? WHERE id = ?", (auto_title, conversation_id))
         conn.commit()
@@ -587,7 +603,7 @@ async def get_status(auth: str = Depends(require_auth)):
         pass
 
     # Conversation count
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute("SELECT COUNT(*) FROM conversations")
     conv_count = c.fetchone()[0]
@@ -623,6 +639,7 @@ async def transcribe_audio(request: Request, audio: UploadFile = File(...), auth
             tmp_path,
             language="en",
             vad_filter=True,
+            beam_size=1,
             initial_prompt=WHISPER_PROMPT
         )
         transcript = " ".join(s.text.strip() for s in segments)
@@ -674,7 +691,7 @@ def transcribe_and_save(tmp_path, timestamp, nextcloud_url, nextcloud_user, next
     nc_auth = (nextcloud_user, nextcloud_password)
     try:
         segments, _ = whisper_model.transcribe(
-            tmp_path, language="en", vad_filter=True, initial_prompt=WHISPER_PROMPT
+            tmp_path, language="en", vad_filter=True, beam_size=1, initial_prompt=WHISPER_PROMPT
         )
         transcript = " ".join(s.text.strip() for s in segments).strip()
         os.unlink(tmp_path)
@@ -760,7 +777,7 @@ async def capture_endpoint(
                     tmp.write(audio_bytes)
                     tmp_audio_path = tmp.name
                 segments, _ = whisper_model.transcribe(
-                    tmp_audio_path, language="en", vad_filter=True, initial_prompt=WHISPER_PROMPT
+                    tmp_audio_path, language="en", vad_filter=True, beam_size=1, initial_prompt=WHISPER_PROMPT
                 )
                 voice_annotation = " ".join(s.text.strip() for s in segments).strip() or None
                 os.unlink(tmp_audio_path)
@@ -969,7 +986,7 @@ async def reindex_status(auth: str = Depends(require_auth)):
 
 @app.delete("/api/conversations")
 async def clear_all_conversations(auth: str = Depends(require_auth)):
-    conn = sqlite3.connect(CONVERSATIONS_DB)
+    conn = _connect_conversations()
     c = conn.cursor()
     c.execute("DELETE FROM messages")
     c.execute("DELETE FROM conversations")

scripts/backup.sh

@@ -6,7 +6,7 @@ mkdir -p "$BACKUP_DIR"
 # Copy critical files
 cp ~/aaronai/memory.md "$BACKUP_DIR/memory-$DATE.md"
 cp ~/aaronai/settings.json "$BACKUP_DIR/settings-$DATE.json"
-cp ~/aaronai/conversations.db "$BACKUP_DIR/conversations-$DATE.db"
+python3 -c "import sqlite3, sys; src = sqlite3.connect('$HOME/aaronai/conversations.db'); dst = sqlite3.connect('$BACKUP_DIR/conversations-$DATE.db'); src.backup(dst); dst.close(); src.close()"
 
 # Keep only last 7 days
 find "$BACKUP_DIR" -name "*.md" -mtime +7 -delete

scripts/dream.py

@@ -16,6 +16,7 @@ import os
 import json
 import sqlite3
 import argparse
+from functools import lru_cache
 from collections import Counter
 from pathlib import Path
 from datetime import datetime, timedelta
@@ -283,6 +284,11 @@ def retrieve_graphiti(mode, task=None, n_results=8, excluded_sources=None):
         print(f"[Graphiti retrieval error: {e}] — falling back to empty.")
         return []
 
+@lru_cache(maxsize=1)
+def _get_embedder():
+    from sentence_transformers import SentenceTransformer
+    return SentenceTransformer("all-MiniLM-L6-v2")
+
 def retrieve(mode, task=None, n_results=8, excluded_sources=None, type_filter=None):
     # E3 experiment: DREAMER_SUBSTRATE=graphiti routes retrieval to Graphiti /search
     # Default behavior: pgvector similarity search (unchanged)
@@ -291,8 +297,7 @@ def retrieve(mode, task=None, n_results=8, excluded_sources=None, type_filter=No
     substrate = os.getenv("DREAMER_SUBSTRATE", "pgvector")
     if substrate == "graphiti":
         return retrieve_graphiti(mode, task=task, n_results=n_results, excluded_sources=excluded_sources)
-    from sentence_transformers import SentenceTransformer
+    embedder = _get_embedder()
-    embedder = SentenceTransformer("all-MiniLM-L6-v2")
     low, high = MODE_RANGES[mode]
 
     if task:
@@ -480,10 +485,11 @@ def write_manifest(date_str, stage_data, corpus_data):
     auth = (NEXTCLOUD_USER, NEXTCLOUD_PASSWORD)
     url = f"{DREAMS_WEBDAV}/dream-manifest-{date_str}.json"
     try:
-        requests.put(url, data=content.encode("utf-8"), auth=auth, timeout=30)
+        response = requests.put(url, data=content.encode("utf-8"), auth=auth, timeout=30)
+        response.raise_for_status()
         print(f"Manifest written: Journal/Dreams/dream-manifest-{date_str}.json")
     except Exception as e:
-        print(f"Manifest write failed (non-critical): {e}")
+        print(f"Manifest write failed — manifest not persisted: {e}")
 
 
 def dream_pipeline(type_filter=None):
@@ -613,8 +619,20 @@ def dream_pipeline(type_filter=None):
     # Write manifest
     all_session_sources = list(session_retrieved)
     all_session_folders = list({extract_folder(s) for s in all_session_sources})
+    total_chunks = 0
+    pg = None
+    try:
+        pg = get_pg()
+        cur = pg.cursor()
+        cur.execute("SELECT COUNT(*) FROM embeddings")
+        total_chunks = cur.fetchone()[0]
+    except Exception as e:
+        print(f"total_chunks query failed (non-critical): {e}")
+    finally:
+        if pg is not None:
+            pg.close()
     corpus_data = {
-        "total_chunks": delta.get("new_chunks", 0),
+        "total_chunks": total_chunks,
         "new_chunks_since_last_dream": delta.get("new_chunks", 0),
         "days_since_last_dream": round(delta.get("days_since_dream", 0), 2),
         "substrate": "pgvector",

scripts/encoding.py

@@ -25,6 +25,30 @@ DEFAULT_CHUNK_SIZE = 500
 DEFAULT_CHUNK_OVERLAP = 50
 
 
+def _docx_cell_paragraphs(cell):
+    yield from (p for p in cell.paragraphs if p.text.strip())
+    for nested in cell.tables:
+        for row in nested.rows:
+            for c in row.cells:
+                yield from _docx_cell_paragraphs(c)
+
+
+def _pptx_shape_text(shape):
+    from pptx.enum.shapes import MSO_SHAPE_TYPE
+    parts = []
+    if shape.shape_type == MSO_SHAPE_TYPE.GROUP:
+        for sub in shape.shapes:
+            parts.extend(_pptx_shape_text(sub))
+        return parts
+    if hasattr(shape, "text") and shape.text.strip():
+        parts.append(shape.text)
+    if getattr(shape, "has_table", False):
+        for cell in shape.table.iter_cells():
+            if cell.text.strip():
+                parts.append(cell.text)
+    return parts
+
+
 def extract_text(filepath: Path) -> str:
     """Return the text of a supported file. Returns "" on any failure or
     unsupported extension. Does not write to ingest_failures — caller decides."""
@@ -32,7 +56,21 @@ def extract_text(filepath: Path) -> str:
     try:
         if suffix == ".docx":
             doc = DocxDocument(filepath)
-            return "\n".join(p.text for p in doc.paragraphs if p.text.strip())
+            parts = [p.text for p in doc.paragraphs if p.text.strip()]
+            for tbl in doc.tables:
+                for row in tbl.rows:
+                    for cell in row.cells:
+                        parts.extend(p.text for p in _docx_cell_paragraphs(cell))
+            for section in doc.sections:
+                parts.extend(p.text for p in section.header.paragraphs if p.text.strip())
+                parts.extend(p.text for p in section.footer.paragraphs if p.text.strip())
+            from docx.oxml.ns import qn
+            for txbx in doc.element.body.findall(".//" + qn("w:txbxContent")):
+                for p in txbx.findall(".//" + qn("w:p")):
+                    text = "".join(t.text or "" for t in p.findall(".//" + qn("w:t")))
+                    if text.strip():
+                        parts.append(text)
+            return "\n".join(parts)
         elif suffix == ".pdf":
             reader = PdfReader(filepath)
             return "".join(
@@ -41,11 +79,15 @@ def extract_text(filepath: Path) -> str:
             )
         elif suffix == ".pptx":
             prs = Presentation(filepath)
-            return "\n".join(
+            parts = []
-                shape.text for slide in prs.slides
+            for slide in prs.slides:
-                for shape in slide.shapes
+                for shape in slide.shapes:
-                if hasattr(shape, "text") and shape.text.strip()
+                    parts.extend(_pptx_shape_text(shape))
-            )
+                if slide.has_notes_slide:
+                    notes = slide.notes_slide.notes_text_frame.text
+                    if notes.strip():
+                        parts.append(notes)
+            return "\n".join(parts)
         elif suffix in {".txt", ".md"}:
             return filepath.read_text(encoding="utf-8", errors="ignore")
     except Exception as e:

scripts/ingest_conversations.py

@@ -18,8 +18,14 @@ CONVERSATIONS_DB = str(Path.home() / "aaronai" / "conversations.db")
 PG_DSN = os.getenv("PG_DSN")
 MIN_EXCHANGES = 3
 
-print("Loading embedding model...")
+_embedder = None
-embedder = SentenceTransformer("all-MiniLM-L6-v2")
+
+def get_embedder():
+    global _embedder
+    if _embedder is None:
+        print("Loading embedding model...")
+        _embedder = SentenceTransformer("all-MiniLM-L6-v2")
+    return _embedder
 
 def get_conversations():
     conn = sqlite3.connect(CONVERSATIONS_DB)
@@ -123,7 +129,7 @@ def run():
         
         # Embed and insert
         texts = [c[1] for c in new_chunks]
-        embeddings = embedder.encode(texts, show_progress_bar=False).tolist()
+        embeddings = get_embedder().encode(texts, show_progress_bar=False).tolist()
         
         for (chunk_id, chunk_text, meta), embedding in zip(new_chunks, embeddings):
             if not meta.get("type"):

scripts/watcher.py

@@ -124,7 +124,7 @@ def resolve_ingest_failure(source: str):
 
 
 def ingest_file(filepath: Path, embedder) -> int:
-    if filepath.name.startswith(("~$", ".")):
+    if filepath.name.startswith(("~$", "~", ".")):
         return 0
     if filepath.suffix.lower() not in SUPPORTED:
         return 0
@@ -168,6 +168,7 @@ def ingest_files(paths: list, embedder, state: dict) -> dict:
     for path in paths:
         count = ingest_file(path, embedder)
         total += count
+        if count > 0:
             state[str(path)] = str(path.stat().st_mtime)
     log.info(f"Ingestion complete. {total} chunks across {len(paths)} files.")
     return state
@@ -196,12 +197,18 @@ def get_changed_files(state: dict) -> list:
             continue
         if path.suffix.lower() not in SUPPORTED:
             continue
-        if path.name.startswith((".", "~$")):
+        if path.name.startswith((".", "~$", "~")):
             continue
         if "Admin/Backups" in str(path) or "Backups" in path.parts:
             continue
         if "Journal/Media" in str(path):
             continue
+        if "Generative Design" in path.parts and "Processing" in path.parts:
+            continue
+        if "Computational Design 2017" in path.parts and "Student Work" in path.parts:
+            continue
+        if path.stat().st_size == 0:
+            continue
         if state.get(str(path)) != str(path.stat().st_mtime):
             changed.append(path)
     return changed
@@ -280,12 +287,16 @@ class IngestHandler(FileSystemEventHandler):
         self.last_event = 0
 
     def _should_ignore(self, path: Path) -> bool:
-        if path.name.startswith((".", "~$")):
+        if path.name.startswith((".", "~$", "~")):
             return True
         if "Admin/Backups" in str(path) or "Backups" in path.parts:
             return True
         if "Journal/Media" in str(path):
             return True
+        if "Generative Design" in path.parts and "Processing" in path.parts:
+            return True
+        if "Computational Design 2017" in path.parts and "Student Work" in path.parts:
+            return True
         return False
 
     def on_created(self, event):