Python Backend & Data Kit

Typed Python services and data work: FastAPI-style backends, pytest, SQL and Postgres design and migrations, data quality checks and tidy notebooks.

A setup for AI coding agents, made by Kitloom. One source, generated for every major agent: Claude Code, OpenAI Codex, Cursor, Google Antigravity, Gemini CLI, GitHub Copilot, Windsurf (Devin), opencode and Aider.

Install

1. Download this repository as a ZIP, or use the Check and download button on its Kitloom page, which checks it first.
2. Read the files. Nothing here runs on its own: the kit has no hooks and no install scripts.
3. Copy the folders and files you want into the root of your project. AGENTS.md is the shared instructions file; edit it to fit your project.

What is inside

Skills

• fastapi-templates — Starting or restructuring a FastAPI service: routers, dependencies, async patterns.
• postgresql-table-design — Designing or reviewing Postgres tables, types, constraints and indexes.
• postgres-database-migration — Planning and testing a schema migration against live data.
• sql-optimization-patterns — A query is slow; read EXPLAIN output and choose indexes.
• schema-exploration — Inspecting an unfamiliar database before writing SQL or answering a data question.
• data-quality-frameworks — Adding validation checks and data contracts to a pipeline or dataset.
• debugging-strategies — Tracking down a bug or flaky test with a systematic approach.
• security-and-hardening — Handling request input, auth, secrets and database access safely.
• observability-and-instrumentation — Adding structured logs, metrics and tracing to a service or job.
• performance-optimization — Profiling and fixing a measured slowdown in code or queries.

Commands

Invoke by name (for example /add-endpoint).

• add-endpoint — Add a typed API endpoint with schemas, tests and error handling.
• explain-query — Analyse a SQL query's plan and suggest safe improvements.
• data-check — Profile a dataset and propose validation checks before it is used downstream.

Sub-agents

• python-reviewer — Delegate to review a Python diff for typing, error handling, security and test gaps.
• sql-reviewer — Delegate to review SQL and migrations for correctness, locking and performance.

Scoped rules

• python-typing — Typing and style habits for every Python change.
• pytest-tests — Applies when writing or editing tests.
• notebooks-sql — Applies to notebooks and SQL files: keep them reproducible and safe.

Where each agent finds it

Agent	Instructions	Skills and commands	Rules	Sub-agents
Claude Code	CLAUDE.md (imports AGENTS.md)	.claude/skills/	.claude/rules/	.claude/agents/
OpenAI Codex	AGENTS.md	.agents/skills/	listed in AGENTS.md	.codex/agents/
Cursor	AGENTS.md	.agents/skills/, .claude/skills/	.cursor/rules/	.claude/agents/
Google Antigravity	AGENTS.md	.agents/skills/	.agents/rules/	.agents/agents/
Gemini CLI	AGENTS.md via .gemini/settings.json	.agents/skills/, .gemini/commands/	listed in AGENTS.md	.gemini/agents/
GitHub Copilot	AGENTS.md	.agents/skills/, .github/prompts/	.github/instructions/	.claude/agents/
Windsurf (Devin)	AGENTS.md	.agents/skills/, .windsurf/workflows/	.windsurf/rules/	—
opencode	AGENTS.md	.agents/skills/, .opencode/commands/	listed in AGENTS.md	.opencode/agents/
Aider	AGENTS.md via .aider.conf.yml	—	—	—

Skills are stored twice (.agents/skills/ and .claude/skills/) because Claude Code reads only its own folder. Agents that read both may list a skill twice; delete the copy you do not need.

Credits

The skills come from these open-source repositories, each under its own licence. Their licence and a notice are kept beside every copied skill.

Skill	Source	Licence
fastapi-templates	wshobson/agents at 46891e7	MIT
postgresql-table-design	wshobson/agents at 46891e7	MIT
postgres-database-migration	timescale/pg-aiguide at 187be00	Apache-2.0
sql-optimization-patterns	wshobson/agents at 46891e7	MIT
schema-exploration	timescale/pg-aiguide at 187be00	Apache-2.0
data-quality-frameworks	wshobson/agents at 46891e7	MIT
debugging-strategies	wshobson/agents at 46891e7	MIT
security-and-hardening	addyosmani/agent-skills at aeac41d	MIT
observability-and-instrumentation	addyosmani/agent-skills at aeac41d	MIT
performance-optimization	addyosmani/agent-skills at aeac41d	MIT

Everything else in this repository (instructions, rules, commands, sub-agents) is by Kitloom under the MIT licence.

Use at your own risk

Read every file before you use it. Agent instructions can make an agent run commands on your machine. This kit was checked with Kitloom's pre-download check, which looks for warning signs but cannot prove anything is safe.