菜单
Use when scraping, cleaning, mapping, or importing the SPPG/DAPODIK data in this repo (script/*.py, script/*.mjs, script/data, script/sql/import_data.sql). Covers the CSV→Postgres flow and known data-quality issues.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when changing the PostgreSQL schema or views for this repo (script/sql/*.sql, tables sppg/dapodik_pd/kecamatan_mapping/kecamatan_coords, views v_sppg/v_sppg_per_kecamatan). Covers ordering, idempotency, and applying against the live sppg-db container.
Use when adding or changing frontend pages/components in this repo (frontend/src, Vue 3 + Vite + Tailwind + Leaflet + Chart.js). Covers routing, API calls, the coverage tier colors, and the build/verify loop.
Use when adding or changing backend API endpoints in this repo (backend/api/routes.js, Express + pg). Covers query conventions, the v_sppg clean view, error handling, and the coverage business rule.
| name | sppg-data-pipeline |
| description | Use when scraping, cleaning, mapping, or importing the SPPG/DAPODIK data in this repo (script/*.py, script/*.mjs, script/data, script/sql/import_data.sql). Covers the CSV→Postgres flow and known data-quality issues. |
Alur: scrape → CSV mentah → clean/remap → CSV bersih → COPY ke Postgres → view analisis.
script/scrape_*.py, script/scrape_*.mjsscript/clean_and_remap.py, script/map_wilayah.py,
script/resume_dapodik_pd.pydocs/reference/ (dapodik_pd_kecamatan.csv, sppg_indonesia_2026.csv)script/data/script/sql/ (setup_db.sql, 02_clean_provinsi.sql, import_data.sql, analysis_views.sql)script/run_pipeline.shdocker exec -i sppg-db psql -U postgres -d sppg < script/sql/setup_db.sql
docker exec -i sppg-db psql -U postgres -d sppg < script/sql/02_clean_provinsi.sql
# import_data.sql meng-COPY dari /tmp -> salin CSV bersih dulu:
docker cp script/data/sppg_clean.csv sppg-db:/tmp/sppg_clean.csv
docker cp script/data/dapodik_clean.csv sppg-db:/tmp/dapodik_clean.csv
docker exec -i sppg-db psql -U postgres -d sppg < script/sql/import_data.sql
docker exec -i sppg-db psql -U postgres -d sppg < script/sql/analysis_views.sql
COPY di import_data.sql sengaja mengikuti urutan header CSV
(...sps, pkbm, skb, sma, smk, slb, sd, smp), bukan urutan definisi tabel. Jangan
"rapikan" tanpa cek header CSV — bisa menggeser data sd/smp ke kolom salah.provinsi di sppg punya ~26 baris kotor (fragmen alamat). Bersih di lapis
query via ref_provinsi + view v_sppg. Perbaikan akar (parser scrape) masih TODO —
catat di docs/business-process.md.kode_kecamatan_bps dibuat otomatis (GENERATED) dari kode_wilayah_bps
(8 digit pertama tanpa titik).Verifikasi jumlah baris & sampel: SELECT count(*) FROM v_sppg; dan cek total_provinsi
di /api/stats tetap 38.