메뉴
NPA property alert system. Scans ALL provider databases (LED, SAM, BAM, JAM, KTB, KBANK) for newly added properties, price drops, best deals, properties near BTS/MRT, and upcoming auctions. Can generate daily reports and run on schedule.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
SOC 직업 분류 기준
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/Arsapol/agent-npa-guy --skill npa-alerts명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
Cross-NPA comparison engine for Thai distressed properties. Finds comparable NPA properties by GPS proximity, computes area benchmarks (median/p25/p75), and returns type-specific comparison results (condo, land, house, commercial). Condos additionally use external market data from market_adapter when available. Non-condo is cross-NPA only — no external market data exists yet.
Run structured analysis (financial, property, data) using the hybrid ralph pattern. Decomposes research questions into parallel investigation tracks with verification.
Team lead skill that decomposes a feature, defines shared contracts, coordinates Agent Teams (creative track) and Ralph loops (mechanical track).
| name | npa-alerts |
| description | NPA property alert system. Scans ALL provider databases (LED, SAM, BAM, JAM, KTB, KBANK) for newly added properties, price drops, best deals, properties near BTS/MRT, and upcoming auctions. Can generate daily reports and run on schedule. |
Automated alert system for NPA property opportunities across all 6 providers.
All commands run from scripts/ directory:
python alert.py new --hours 24 # Last 24h across all providers
python alert.py new --hours 48 --source bam # BAM only, last 48h
python alert.py new --source jam # JAM only
python alert.py new --source ktb # KTB only
python alert.py new --source kbank # KBANK only
python alert.py new --json # JSON output
Sources: all (default), led, sam, bam, jam, ktb, kbank
python alert.py drops --hours 24 --min-drop 5 # ≥5% price drops in last 24h
python alert.py drops --hours 168 --min-drop 10 --limit 20 # ≥10% drops in last 7 days
python alert.py drops --json # JSON output
Sources: SAM (sam_price_history), BAM, JAM, KTB, KBANK (all via price_history tables)
python alert.py increases --hours 24 --min-increase 5 # ≥5% price increases in last 24h
python alert.py increases --hours 168 --min-increase 10 # ≥10% increases in last 7 days
python alert.py increases --json # JSON output
Why it matters: NPA prices normally only go DOWN. A price increase signals something changed — re-appraisal, encumbrance removed, market shift, or area gentrification. Worth investigating the reason.
python alert.py deals --min-discount 30 --limit 20 # >30% below appraisal
python alert.py deals --max-price 5000000 # Under ฿5M
python alert.py bts --meters 500 --limit 20 # Within 500m of any station
python alert.py bts --meters 1000 --max-price 3000000 # Under ฿3M within 1km
python alert.py bts --source bam # BAM only near transit
Sources: all (default), sam, bam, jam, ktb, kbank
(LED has no GPS data)
python alert.py upcoming --days 14 # Next 2 weeks
python alert.py upcoming --days 30 --province "กรุงเทพ" # Bangkok only
python alert.py report # All sections combined
| Section | Sources | Description |
|---|---|---|
| New Properties | LED + SAM + BAM + JAM + KTB + KBANK | Properties added in last 24h (by first_seen_at / created_at) |
| Price Drops | SAM + BAM + JAM + KTB + KBANK | Price drops ≥5% detected via price_history tables |
| Price Increases | SAM + BAM + JAM + KTB + KBANK | Price increases ≥5% — unusual for NPA, investigate why |
| Best Deals | LED | Properties >30% below enforcement officer appraisal |
| Near BTS/MRT | SAM + BAM + JAM + KTB + KBANK | Properties within 500m of transit (has GPS) |
| Upcoming Auctions | LED | Properties with auction dates in next 30 days |
| Provider | Count | Price Field | GPS | Price History | Notes |
|---|---|---|---|---|---|
| LED (กรมบังคับคดี) | 17,705 | satang (÷100) | ❌ | — | Court auctions, created_at |
| SAM (บสส.) | 4,707 | baht | ✅ | ✅ sam_price_history | Best data quality, first_seen_at |
| BAM (BAY) | 6,798 | baht (discount_price) | ✅ | ✅ bam_price_history | Grade A/B properties, first_seen_at |
| JAM (IMC) | 38,820 | baht (discount) | ✅ | ✅ jam_price_history | 2nd largest NPA source, first_seen_at |
| KTB (Krungthai) | 2,671 | baht (price) | ✅ | ✅ ktb_price_history | Biggest discounts, first_seen_at |
| KBANK | 13,361 | baht (promotion_price) | ✅ | ✅ kbank_price_history | Thin condo inventory, first_seen_at |
Hardcoded BTS/MRT stations for distance calculations. Covers:
next_auction_date is stored as varchar (not date) — queries cast to dateenforcement_officer_price_satang) serves as the appraisal baseline for LED dealschange_type = 'price_drop' in price_history tables (recorded by scrapers)