# specsfy-aux-database

Maintains .specsfy/DATABASE.md with tables of databases, schemas, fields, relations and migrations so the data map matches the code.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 437 installs · ★ 87
- Author: promovaweb
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/promovaweb/specsfy/tree/HEAD/skills/specsfy-aux-database/SKILL.md
- Updated: 2026-10-20
- HTML: https://affpapa.org/en/nevibecode/skills/specsfy-aux-database

## What it does

The skill walks schemas, migrations, persistent models and configs, then writes a tabular inventory into `.specsfy/DATABASE.md`. Markdown tables are there for a reason: they are easy to search and diff. Relations, indexes, constraints, ownership, retention and purpose that cannot be inferred safely get flagged as unknown instead of guessed.

Load `$specsfy-setup` at the project root first and stop if it reports an open item. The skill then reads local instructions, `PROJECT.md`, `STACK.md` and the whole `DATABASE.md`, runs `node scripts/update_database.mjs --project <raiz>` and checks the detected inventory against executable sources. Anything outside the `specsfy:database` block stays untouched, so your notes and decisions survive the update.

It runs without asking questions: missing details become a marker for the next step rather than a prompt. A schema change counts as unfinished until `DATABASE.md` reflects the new structure and the relevant tests pass. Connection strings, passwords, tokens and `.env` values never make it into the file, which matters when the docs live in a shared repository.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add promovaweb/specsfy --skill specsfy-aux-database -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/specsfy-aux-database/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/specsfy-aux-database/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add promovaweb/specsfy --skill specsfy-aux-database -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/specsfy-aux-database/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add promovaweb/specsfy --skill specsfy-aux-database -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/specsfy-aux-database/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install promovaweb/specsfy specsfy-aux-database
```

2. **Manually: put the files in place**
- File: `.github/skills/specsfy-aux-database/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add promovaweb/specsfy --skill specsfy-aux-database -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/promovaweb/specsfy --path skills/specsfy-aux-database
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/specsfy-aux-database/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add promovaweb/specsfy --skill specsfy-aux-database -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/promovaweb/specsfy/tree/HEAD/skills/specsfy-aux-database/SKILL.md

## When to use it
- build a tabular inventory of databases, schemas and fields from code
- check whether the data map still matches current migrations
- flag relations, indexes and constraints that cannot be inferred
- refresh `DATABASE.md` without wiping team notes and decisions
- treat a schema change as done only after tests pass

## Example prompts
- update the data map in `.specsfy/DATABASE.md`
- check the database inventory against the current migrations
- I added a table and a migration — refresh the persistence docs

## Similar
- [database-designer](https://affpapa.org/en/nevibecode/skills/database-designer.md) — skill · 819 installs
- [metabase-database-metadata](https://affpapa.org/en/nevibecode/skills/metabase-database-metadata.md) — skill · 560 installs
- [database-schema](https://affpapa.org/en/nevibecode/skills/database-schema.md) — skill · 302 installs
- [convex-migrate](https://affpapa.org/en/nevibecode/skills/convex-migrate.md) — skill · 36100 installs
- [convex-migrate-rehearse](https://affpapa.org/en/nevibecode/skills/convex-migrate-rehearse.md) — skill · 35900 installs
- [nosql-database-design](https://affpapa.org/en/nevibecode/skills/nosql-database-design.md) — skill · 486 installs

From the same set: [promovaweb/specsfy](https://affpapa.org/en/nevibecode/by/promovaweb/specsfy.md) — 3 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
