# Python Database

> Implement Python database access with parameterized SQL, transaction scope, connection helpers, and repository seams. Use when editing Postgres queries, repositories, transactions, pooling, or persistence boundaries in Python.

- Skill: `hoangnguyen0403/python-database` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add hoangnguyen0403/python-database`
- Raw SKILL.md: https://api.skillmd.com/api/skills/hoangnguyen0403/python-database/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Data & Analytics
- Author: HoangNguyen0403 (https://skillmd.com/u/hoangnguyen0403)
- Updated: 2026-08-19
- Page: https://skillmd.com/skills/hoangnguyen0403/python-database

---


# Python Database

## **Priority: P1 (HIGH)**

## Rules

- Parameterize every query; never interpolate user or runtime data into SQL strings.
- Keep connection and transaction ownership explicit.
- Hide storage shape behind repository or query helpers when business logic depends on it.
- Normalize JSON and null handling at the persistence boundary.

## Recipe

1. **Open connections through one helper or pool abstraction**.
2. **Use context managers** for connection and transaction lifetime.
3. **Keep one business action in one transaction**.
4. **Return typed rows or normalized dicts**, not raw cursor tuples leaking everywhere.
5. **Test malformed/null metadata paths** when patching JSON payloads.

## Anti-Patterns

- **No f-string SQL**.
- **No transaction split-brain** across handler, service, and repo.
- **No unbounded connection churn** in loops.
- **No DB truth hidden behind report formatting code**.

## References

- [Framework Map](../references/framework-map.md)
- [DB Boundaries](references/db-boundaries.md)