Initial commit: clean project structure

- Backend: FastAPI + SQLAlchemy + Celery (Python 3.11+)
- Frontend: Vue 3 + TypeScript + Pinia + Tailwind
- Admin Frontend: separate Vue 3 app for management
- Docker Compose: 9 services orchestration
- Specs: design prototypes, memory system PRD, product roadmap

Cleanup performed:
- Removed temporary debug scripts from backend root
- Removed deprecated admin_app.py (embedded UI)
- Removed duplicate docs from admin-frontend
- Updated .gitignore for Vite cache and egg-info

backend/app/services/achievement_extractor.py

@@ -0,0 +1,85 @@
+"""Achievement extraction service."""
+
+import json
+import re
+
+import httpx
+
+from app.core.config import settings
+from app.core.logging import get_logger
+from app.core.prompts import ACHIEVEMENT_EXTRACTION_PROMPT
+
+logger = get_logger(__name__)
+
+TEXT_API_BASE = "https://generativelanguage.googleapis.com/v1beta/models"
+
+
+async def extract_achievements(story_text: str) -> list[dict]:
+    """Extract achievements from story text using LLM."""
+    if not settings.text_api_key:
+        logger.warning("achievement_extraction_skipped", reason="missing_text_api_key")
+        return []
+
+    model = settings.text_model or "gemini-2.0-flash"
+    url = f"{TEXT_API_BASE}/{model}:generateContent"
+
+    prompt = ACHIEVEMENT_EXTRACTION_PROMPT.format(story_text=story_text)
+    payload = {
+        "contents": [{"parts": [{"text": prompt}]}],
+        "generationConfig": {
+            "responseMimeType": "application/json",
+            "temperature": 0.2,
+            "topP": 0.9,
+        },
+    }
+
+    async with httpx.AsyncClient(timeout=30) as client:
+        response = await client.post(
+            url,
+            json=payload,
+            headers={"x-goog-api-key": settings.text_api_key},
+        )
+        response.raise_for_status()
+        result = response.json()
+
+    candidates = result.get("candidates") or []
+    if not candidates:
+        logger.warning("achievement_extraction_empty")
+        return []
+
+    parts = candidates[0].get("content", {}).get("parts") or []
+    if not parts or "text" not in parts[0]:
+        logger.warning("achievement_extraction_missing_text")
+        return []
+
+    response_text = parts[0]["text"]
+    clean_json = response_text
+    if response_text.startswith("```json"):
+        clean_json = re.sub(r"^```json\n|```$", "", response_text)
+
+    try:
+        parsed = json.loads(clean_json)
+    except json.JSONDecodeError:
+        logger.warning("achievement_extraction_parse_failed")
+        return []
+
+    achievements = parsed.get("achievements")
+    if not isinstance(achievements, list):
+        return []
+
+    normalized: list[dict] = []
+    for item in achievements:
+        if not isinstance(item, dict):
+            continue
+        a_type = str(item.get("type", "")).strip()
+        description = str(item.get("description", "")).strip()
+        score = item.get("score", 0)
+        if not a_type or not description:
+            continue
+        normalized.append({
+            "type": a_type,
+            "description": description,
+            "score": score
+        })
+
+    return normalized