wip

.agents/skills/prisma-cli/references/agent-safety.md

@@ -0,0 +1,27 @@
+# AI safety checkpoint for destructive commands
+
+Prisma detects common AI-agent environments and blocks these commands until the user gives explicit consent:
+
+- `prisma migrate reset`
+- `prisma db push --force-reset`
+- `prisma db push --accept-data-loss`
+
+## Required workflow
+
+1. Inspect the target database/config and explain exactly what can be deleted or reset.
+2. Ask the user for explicit consent immediately before the action.
+3. Run the command only after that consent.
+
+For an agent-run subprocess, Prisma accepts the exact consent text through:
+
+```bash
+PRISMA_USER_CONSENT_FOR_DANGEROUS_AI_ACTION='<exact user consent message>' prisma migrate reset --force
+```
+
+The value must match the user's message exactly and must not contain added quotes or newlines. Never fabricate consent, reuse an old unrelated approval, or bypass the checkpoint by hiding agent-detection environment variables.
+
+The MCP server has no `migrate-reset` tool. Use the shell command only after consent.
+
+## Reference
+
+- [Prisma ORM 7.9.0 release](https://github.com/prisma/prisma/releases/tag/7.9.0)

.agents/skills/prisma-cli/references/complete.md

@@ -0,0 +1,22 @@
+# prisma complete
+
+Prints a shell completion script.
+
+```bash
+prisma complete zsh
+prisma complete bash
+prisma complete fish
+prisma complete powershell
+```
+
+For a direct global CLI installation, load the output using the shell's normal startup mechanism. For example, in zsh:
+
+```bash
+source <(prisma complete zsh)
+```
+
+Prisma also integrates with supported package-manager completion flows. `npx` and `bunx` do not themselves provide completion; invoke the installed binary or the package manager's supported execution form such as `npm exec` or `bun x`.
+
+## Reference
+
+- [Prisma ORM 7.9.0 release](https://github.com/prisma/prisma/releases/tag/7.9.0)

.agents/skills/prisma-cli/references/db-execute.md

@@ -0,0 +1,78 @@
+# prisma db execute
+
+Execute native commands (SQL) to your database.
+
+## Command
+
+```bash
+prisma db execute [options]
+```
+
+## What It Does
+
+- Connects to your database using the configured datasource
+- Executes a script provided via file (`--file`) or stdin (`--stdin`)
+- Useful for running raw SQL, maintenance tasks, or applying diffs from `migrate diff`
+- Not supported on MongoDB
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--file` | Path to a file containing the script to execute |
+| `--stdin` | Use terminal standard input as the script |
+| `--config` | Custom path to your Prisma config file |
+
+## Current Option Surface
+
+`prisma db execute` uses the datasource configured in `prisma.config.ts`. Use `--config` if you need a separate config file for another environment.
+
+## Examples
+
+### Execute from file
+
+```bash
+prisma db execute --file ./script.sql
+```
+
+### Execute from stdin
+
+```bash
+echo "TRUNCATE TABLE User;" | prisma db execute --stdin
+```
+
+### Execute `migrate diff` output
+
+Pipe the output of `migrate diff` directly to the database:
+
+```bash
+prisma migrate diff \
+  --from-empty \
+  --to-schema prisma/schema.prisma \
+  --script \
+| prisma db execute --stdin
+```
+
+## Configuration
+
+Uses `datasource` from `prisma.config.ts`:
+
+```typescript
+export default defineConfig({
+  datasource: {
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+## Use Cases
+
+- **Manual Migrations**: Applying raw SQL changes
+- **Data Maintenance**: Truncating tables, cleaning up data
+- **Schema Synchronization**: Applying `migrate diff` scripts
+- **Debugging**: Running test queries (though typically not for fetching data)
+
+## Limitations
+
+- **No Data Return**: The command reports success/failure, not query results (rows). Use Prisma Client or `prisma studio` to view data.
+- **SQL Only**: Primarily for SQL databases.

.agents/skills/prisma-cli/references/db-pull.md

@@ -0,0 +1,185 @@
+# prisma db pull
+
+Introspects an existing database and updates your Prisma schema to reflect its structure.
+
+## Command
+
+```bash
+prisma db pull [options]
+```
+
+## What It Does
+
+- Connects to your database
+- Reads the database schema (tables, columns, relations, indexes)
+- Updates `schema.prisma` with corresponding Prisma models
+- For MongoDB, samples data to infer schema
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--force` | Ignore current Prisma schema file |
+| `--print` | Print the introspected Prisma schema to stdout |
+| `--schema` | Custom path to your Prisma schema |
+| `--config` | Custom path to your Prisma config file |
+| `--url` | Override the datasource URL from the Prisma config file |
+| `--composite-type-depth` | Specify the depth for introspecting composite types (default: -1 for infinite, 0 = off) |
+| `--schemas` | Specify the database schemas to introspect |
+| `--local-d1` | Generate a Prisma schema from a local Cloudflare D1 database |
+
+## Examples
+
+### Basic introspection
+
+```bash
+prisma db pull
+```
+
+### Preview without writing
+
+```bash
+prisma db pull --print
+```
+
+Outputs schema to terminal for review.
+
+### Force overwrite
+
+```bash
+prisma db pull --force
+```
+
+Replaces schema file, losing any manual customizations.
+
+## Prerequisites
+
+Configure database connection in `prisma.config.ts`:
+
+```typescript
+import 'dotenv/config'
+import { defineConfig, env } from 'prisma/config'
+
+export default defineConfig({
+  schema: 'prisma/schema.prisma',
+  datasource: {
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+## Workflow
+
+### Starting from existing database
+
+1. Initialize Prisma:
+   ```bash
+   prisma init
+   ```
+
+2. Configure database URL
+
+3. Pull schema:
+   ```bash
+   prisma db pull
+   ```
+
+4. Review and customize generated schema
+
+5. Generate client:
+   ```bash
+   prisma generate
+   ```
+
+### Syncing changes from database
+
+When database changes are made outside Prisma:
+
+```bash
+prisma db pull
+prisma generate
+```
+
+## Generated Schema Example
+
+Database tables become Prisma models:
+
+```sql
+-- Database tables
+CREATE TABLE users (
+  id SERIAL PRIMARY KEY,
+  email VARCHAR(255) UNIQUE NOT NULL,
+  name VARCHAR(100)
+);
+
+CREATE TABLE posts (
+  id SERIAL PRIMARY KEY,
+  title VARCHAR(255) NOT NULL,
+  author_id INTEGER REFERENCES users(id)
+);
+```
+
+Becomes:
+
+```prisma
+model users {
+  id    Int     @id @default(autoincrement())
+  email String  @unique @db.VarChar(255)
+  name  String? @db.VarChar(100)
+  posts posts[]
+}
+
+model posts {
+  id        Int    @id @default(autoincrement())
+  title     String @db.VarChar(255)
+  author_id Int?
+  users     users? @relation(fields: [author_id], references: [id])
+}
+```
+
+## Post-Introspection Cleanup
+
+After `db pull`, consider:
+
+1. **Rename models** to PascalCase:
+   ```prisma
+   model User {  // Was: users
+     @@map("users")
+   }
+   ```
+
+2. **Rename fields** to camelCase:
+   ```prisma
+   authorId Int? @map("author_id")
+   ```
+
+3. **Add relation names** for clarity:
+   ```prisma
+   author User? @relation("PostAuthor", fields: [authorId], references: [id])
+   ```
+
+4. **Add documentation**:
+   ```prisma
+   /// User account information
+   model User {
+     /// Primary email for authentication
+     email String @unique
+   }
+   ```
+
+## MongoDB Introspection
+
+For MongoDB, `db pull` samples documents to infer schema:
+
+```bash
+prisma db pull
+```
+
+May require manual refinement since MongoDB is schemaless.
+
+## Warning
+
+`db pull` overwrites your schema file. Always:
+- Commit current schema before pulling
+- Use `--print` to preview first
+- Backup customizations you want to keep

.agents/skills/prisma-cli/references/db-push.md

@@ -0,0 +1,150 @@
+# prisma db push
+
+Pushes schema changes directly to database without creating migrations. Ideal for prototyping.
+
+## Command
+
+```bash
+prisma db push [options]
+```
+
+## What It Does
+
+- Syncs your Prisma schema to the database
+- Creates database if it doesn't exist
+- Does NOT create migration files
+- Does NOT track migration history
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--force-reset` | Force a reset of the database before push |
+| `--accept-data-loss` | Ignore data loss warnings |
+| `--schema` | Custom path to your Prisma schema |
+| `--config` | Custom path to your Prisma config file |
+| `--url` | Override the datasource URL from the Prisma config file |
+
+When Prisma detects an AI agent, `--force-reset` and `--accept-data-loss` require explicit user consent. Follow `agent-safety.md`; never infer or fabricate the consent text.
+
+### Follow-up Command
+
+- Run `prisma generate` explicitly when you need refreshed client output
+
+## Examples
+
+### Basic push
+
+```bash
+prisma db push
+```
+
+### Accept data loss
+
+```bash
+prisma db push --accept-data-loss
+```
+
+Required when changes would delete data (dropping columns, etc.)
+
+### Force reset
+
+```bash
+prisma db push --force-reset
+```
+
+Completely resets database and applies schema.
+
+### Full workflow
+
+```bash
+prisma db push
+prisma generate
+```
+
+## When to Use
+
+- **Prototyping** - Rapid schema iteration
+- **Local development** - Quick schema changes
+- **MongoDB** - Primary workflow (migrations not supported)
+- **Testing** - Setting up test databases
+
+## When NOT to Use
+
+- **Production** - Use `migrate deploy`
+- **Team collaboration** - Use migrations for trackable changes
+- **When you need rollback** - Migrations provide history
+
+## Comparison with migrate dev
+
+| Feature | db push | migrate dev |
+|---------|---------|-------------|
+| Creates migration files | No | Yes |
+| Tracks history | No | Yes |
+| Requires shadow database | No | Yes |
+| Speed | Faster | Slower |
+| Rollback capability | No | Yes |
+| Best for | Prototyping | Development |
+
+## MongoDB Workflow
+
+MongoDB doesn't support migrations. Use `db push` exclusively:
+
+```bash
+# Schema changes for MongoDB
+prisma db push
+prisma generate
+```
+
+## Common Patterns
+
+### Prototyping workflow
+
+```bash
+# Make schema changes
+# ...
+
+# Push to database
+prisma db push
+
+# Generate client
+prisma generate
+
+# Test your changes
+# Repeat as needed
+```
+
+### Reset and start fresh
+
+```bash
+prisma db push --force-reset
+prisma db seed
+```
+
+### Handling conflicts
+
+If `db push` can't apply changes safely:
+
+```
+Error: The following changes cannot be applied:
+  - Removing field `email` would cause data loss
+  
+Use --accept-data-loss to proceed
+```
+
+Decide whether data loss is acceptable, then:
+
+```bash
+prisma db push --accept-data-loss
+```
+
+## Transition to Migrations
+
+When ready for production, switch to migrations:
+
+```bash
+# Create baseline migration from current schema
+prisma migrate dev --name init
+```
+
+Then use `migrate dev` for future changes.

.agents/skills/prisma-cli/references/db-seed.md

@@ -0,0 +1,188 @@
+# prisma db seed
+
+Runs your database seed script to populate data.
+
+## Command
+
+```bash
+prisma db seed [options]
+```
+
+## What It Does
+
+- Executes your configured seed script
+- Populates database with initial/test data
+- Runs independently (not auto-run by migrations in v7)
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--config` | Custom path to your Prisma config file |
+| `--` | Pass custom arguments to seed script |
+
+## Configuration
+
+Configure seed script in `prisma.config.ts`:
+
+```typescript
+import 'dotenv/config'
+import { defineConfig, env } from 'prisma/config'
+
+export default defineConfig({
+  schema: 'prisma/schema.prisma',
+  migrations: {
+    path: 'prisma/migrations',
+    seed: 'tsx prisma/seed.ts',  // Your seed command
+  },
+  datasource: {
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+### Common seed commands
+
+```typescript
+// TypeScript with tsx
+seed: 'tsx prisma/seed.ts'
+
+// TypeScript with ts-node
+seed: 'ts-node prisma/seed.ts'
+
+// JavaScript
+seed: 'node prisma/seed.js'
+```
+
+## Seed Script Example
+
+```typescript
+// prisma/seed.ts
+import { PrismaClient } from '../generated/client'
+
+const prisma = new PrismaClient()
+
+async function main() {
+  // Create users
+  const alice = await prisma.user.upsert({
+    where: { email: 'alice@prisma.io' },
+    update: {},
+    create: {
+      email: 'alice@prisma.io',
+      name: 'Alice',
+      posts: {
+        create: {
+          title: 'Hello World',
+          published: true,
+        },
+      },
+    },
+  })
+
+  const bob = await prisma.user.upsert({
+    where: { email: 'bob@prisma.io' },
+    update: {},
+    create: {
+      email: 'bob@prisma.io',
+      name: 'Bob',
+    },
+  })
+
+  console.log({ alice, bob })
+}
+
+main()
+  .then(async () => {
+    await prisma.$disconnect()
+  })
+  .catch(async (e) => {
+    console.error(e)
+    await prisma.$disconnect()
+    process.exit(1)
+  })
+```
+
+## Examples
+
+### Run seed
+
+```bash
+prisma db seed
+```
+
+### With custom arguments
+
+```bash
+prisma db seed -- --environment development
+```
+
+Arguments after `--` are passed to your seed script.
+
+## Current Workflow
+
+Run seeding explicitly after migrations when you need seed data:
+
+```bash
+prisma migrate dev --name init
+prisma generate
+prisma db seed  # Must run explicitly
+```
+
+## Idempotent Seeding
+
+Use `upsert` to make seeds re-runnable:
+
+```typescript
+// Good: Can run multiple times
+await prisma.user.upsert({
+  where: { email: 'alice@prisma.io' },
+  update: {},  // Don't change existing
+  create: { email: 'alice@prisma.io', name: 'Alice' },
+})
+
+// Bad: Fails on second run
+await prisma.user.create({
+  data: { email: 'alice@prisma.io', name: 'Alice' },
+})
+```
+
+## Common Patterns
+
+### Development reset
+
+```bash
+prisma migrate reset --force
+prisma db seed
+```
+
+### Conditional seeding
+
+```typescript
+// prisma/seed.ts
+const count = await prisma.user.count()
+if (count === 0) {
+  // Only seed if empty
+  await seedUsers()
+}
+```
+
+### Environment-specific seeds
+
+```typescript
+// prisma/seed.ts
+const env = process.env.NODE_ENV || 'development'
+
+if (env === 'development') {
+  await seedDevData()
+} else if (env === 'test') {
+  await seedTestData()
+}
+```
+
+## Best Practices
+
+1. Use `upsert` for idempotent seeds
+2. Keep seeds focused and minimal
+3. Use realistic but fake data
+4. Document required seed data
+5. Version control your seed scripts

.agents/skills/prisma-cli/references/debug.md

@@ -0,0 +1,46 @@
+# prisma debug
+
+Prints information helpful for debugging and bug reports.
+
+## Command
+
+```bash
+prisma debug [options]
+```
+
+## What It Does
+
+Outputs details about your Prisma environment, including:
+- Prisma CLI version
+- Prisma Client version (if installed)
+- Engine binaries (Query Engine, Migration Engine, etc.)
+- Platform information (OS, Architecture)
+- Node.js version
+- Configured datasource provider
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Example Output
+
+```
+prisma               : 7.3.0
+@prisma/client       : 7.3.0
+Operating System     : darwin
+Architecture         : arm64
+Node.js              : v20.10.0
+TypeScript           : 5.3.3
+Query Compiler       : enabled
+PSL                  : ...
+Schema Engine        : ...
+```
+
+## When to Use
+
+- **Troubleshooting**: Checking version mismatches
+- **Reporting Issues**: Including environment info in GitHub issues
+- **Verifying Installation**: Ensuring correct binaries are downloaded

.agents/skills/prisma-cli/references/dev.md

@@ -0,0 +1,157 @@
+# prisma dev
+
+Starts a local Prisma Postgres database for development. Provides a PostgreSQL-compatible database that runs entirely on your machine.
+
+## Command
+
+```bash
+prisma dev [options]
+```
+
+## What It Does
+
+- Starts a local PostgreSQL-compatible database
+- Runs in your terminal or as a background process
+- Perfect for development and testing
+- Easy migration to Prisma Postgres cloud in production
+
+## Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `--name` / `-n` | Name for the database instance | `default` |
+| `--port` / `-p` | HTTP server port | `51213` |
+| `--db-port` / `-P` | Database server port | `51214` |
+| `--shadow-db-port` | Shadow database port (for migrations) | `51215` |
+| `--detach` / `-d` | Run in background | `false` |
+| `--debug` | Enable debug logging | `false` |
+
+## Examples
+
+### Start local database
+
+```bash
+prisma dev
+```
+
+Interactive mode with keyboard shortcuts:
+- `q` - Quit
+- `h` - Show HTTP URL  
+- `t` - Show TCP URLs
+
+### Named instance
+
+```bash
+prisma dev --name myproject
+```
+
+Useful for multiple projects.
+
+### Background mode
+
+```bash
+prisma dev --detach
+```
+
+Frees your terminal for other commands.
+
+### Custom ports
+
+```bash
+prisma dev --port 5000 --db-port 5432
+```
+
+## Instance Management
+
+### List all instances
+
+```bash
+prisma dev ls
+```
+
+Shows all local Prisma Postgres instances with status.
+
+### Start existing instance
+
+```bash
+prisma dev start myproject
+```
+
+Starts a previously created instance in background.
+
+### Stop instance
+
+```bash
+prisma dev stop myproject
+```
+
+### Stop with glob pattern
+
+```bash
+prisma dev stop "myproject*"
+```
+
+Stops all instances matching pattern.
+
+### Remove instance
+
+```bash
+prisma dev rm myproject
+```
+
+Removes instance data from filesystem.
+
+### Force remove (stops first)
+
+```bash
+prisma dev rm myproject --force
+```
+
+## Configuration
+
+Configure your `prisma.config.ts` to use local Prisma Postgres:
+
+```typescript
+import 'dotenv/config'
+import { defineConfig, env } from 'prisma/config'
+
+export default defineConfig({
+  schema: 'prisma/schema.prisma',
+  migrations: {
+    path: 'prisma/migrations',
+  },
+  datasource: {
+    // Local Prisma Postgres URL (from prisma dev output)
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+## Workflow
+
+1. Start local database:
+   ```bash
+   prisma dev
+   ```
+
+2. In another terminal, run migrations:
+   ```bash
+   prisma migrate dev
+   ```
+
+3. Generate client:
+   ```bash
+   prisma generate
+   ```
+
+4. Run your application
+
+## Production Migration
+
+When ready for production, switch to Prisma Postgres cloud:
+
+```bash
+prisma init --db
+```
+
+Update your `DATABASE_URL` to the cloud connection string.

.agents/skills/prisma-cli/references/format.md

@@ -0,0 +1,48 @@
+# prisma format
+
+Formats your Prisma schema file.
+
+## Command
+
+```bash
+prisma format [options]
+```
+
+## What It Does
+
+- Fixes formatting (indentation, spacing)
+- Adds missing back-relations (e.g., adds the other side of a relation)
+- Adds missing relation arguments (e.g., `fields`, `references`)
+- Sorts fields and attributes (opinionated)
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Examples
+
+### Format default schema
+
+```bash
+prisma format
+```
+
+### Format specific schema
+
+```bash
+prisma format --schema=./custom/schema.prisma
+```
+
+## Behavior
+
+`prisma format` modifies the file in place. It is equivalent to "Prettier for Prisma schemas" but also has semantic understanding to fix/add missing schema definitions.
+
+## Use in Editor
+
+Most Prisma editor extensions (VS Code, WebStorm) run `prisma format` automatically on save. This command is useful for:
+- CI pipelines (check formatting)
+- CLI-based workflows
+- Fixing large schema refactors

.agents/skills/prisma-cli/references/generate.md

@@ -0,0 +1,173 @@
+# prisma generate
+
+Generates assets based on the generator blocks in your Prisma schema, most commonly Prisma Client.
+
+## Command
+
+```bash
+prisma generate [options]
+```
+
+## Bun Runtime
+
+If you're using Bun, run Prisma with `bunx --bun` so it doesn't fall back to Node.js:
+
+```bash
+bunx --bun prisma generate
+```
+
+## What It Does
+
+1. Reads your `schema.prisma` file
+2. Generates a customized Prisma Client based on your models
+3. Outputs to the directory specified in the generator block
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Custom path to your Prisma schema |
+| `--config` | Custom path to your Prisma config file |
+| `--sql` | Generate typed sql module |
+| `--watch` | Watch the Prisma schema and rerun after a change |
+| `--generator` | Generator to use (may be provided multiple times) |
+| `--no-hints` | Hides the hint messages but still outputs errors and warnings |
+| `--require-models` | Do not allow generating a client without models |
+
+## Examples
+
+### Basic generation
+
+```bash
+prisma generate
+```
+
+### Watch mode (development)
+
+```bash
+prisma generate --watch
+```
+
+Auto-regenerates when `schema.prisma` changes.
+
+### Specific generator
+
+```bash
+prisma generate --generator client
+```
+
+### Multiple generators
+
+```bash
+prisma generate --generator client --generator zod_schemas
+```
+
+### Typed SQL generation
+
+```bash
+prisma generate --sql
+```
+
+## Schema Configuration
+
+```prisma
+generator client {
+  provider = "prisma-client"
+  output   = "../generated"
+}
+```
+
+### Current Generator Behavior
+
+- `prisma-client` is the standard generator
+- `output` is required when using `prisma-client`
+- `prisma-client` supports both ESM and CommonJS via `moduleFormat`
+- `compilerBuild` supports `fast` and `small` query compiler artifacts
+- Use TypeScript `satisfies` for typed query fragments with `prisma-client`
+- Import Prisma Client from your generated output path, for example:
+
+```typescript
+import { PrismaClient } from '../generated/prisma/client'
+```
+
+### Compiler Build Tuning
+
+Use `compilerBuild` when you need to trade artifact size against the default build:
+
+```prisma
+generator client {
+  provider      = "prisma-client"
+  output        = "../generated"
+  compilerBuild = "small"
+}
+```
+
+- `fast` is the default build for most targets
+- `small` is useful for size-constrained targets
+- Prisma defaults `vercel-edge` targets to `small`
+
+## Common Patterns
+
+### After schema changes
+
+```bash
+prisma migrate dev --name my_migration
+prisma generate
+```
+
+Run `prisma generate` whenever you need refreshed client code after schema-changing commands.
+
+### CI/CD pipeline
+
+```bash
+prisma generate
+```
+
+Run before building your application.
+
+### Multiple generators
+
+```prisma
+generator client {
+  provider = "prisma-client"
+  output   = "../generated"
+}
+
+generator zod {
+  provider = "zod-prisma-types"
+  output   = "../generated/zod"
+}
+```
+
+```bash
+prisma generate  # Runs all generators
+```
+
+## Output Structure
+
+After running `prisma generate`, your output directory contains:
+
+```
+generated/
+├── browser.ts
+├── client.ts
+├── commonInputTypes.ts
+├── models/
+├── enums.ts
+├── models.ts
+└── ...
+```
+
+Import the client:
+
+```typescript
+import { PrismaClient, Prisma } from '../generated/prisma/client'
+```
+
+Import browser-safe types:
+
+```typescript
+import { Prisma } from '../generated/prisma/browser'
+import { Role } from '../generated/prisma/enums'
+import type { UserModel } from '../generated/prisma/models/User'
+```

.agents/skills/prisma-cli/references/init.md

@@ -0,0 +1,139 @@
+# prisma init
+
+Bootstraps a fresh Prisma ORM project in the current directory.
+
+## Command
+
+```bash
+prisma init [options]
+```
+
+## Bun Runtime
+
+If you're using Bun, run Prisma with `bunx --bun` so it doesn't fall back to Node.js:
+
+```bash
+bunx --bun prisma init
+```
+
+## What It Creates
+
+- `prisma/schema.prisma` - Your Prisma schema file
+- `prisma.config.ts` - TypeScript configuration for Prisma CLI
+- `.env` - Environment variables (DATABASE_URL)
+- `.gitignore` - Ensures `.env` is ignored and appends the generated client path
+
+## Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `--datasource-provider` | Database provider: `postgresql`, `mysql`, `sqlite`, `sqlserver`, `mongodb`, `cockroachdb` | `postgresql` |
+| `--db` | Provisions a fully managed Prisma Postgres database on the Prisma Data Platform | - |
+| `--url` | Define a custom datasource url | - |
+| `--generator-provider` | Define the generator provider to use | `prisma-client` |
+| `--output` | Define Prisma Client generator output path to use | - |
+| `--preview-feature` | Define a preview feature to use | - |
+| `--with-model` | Add example model to created schema file | - |
+| `--no-skills` | Skip the best-effort installation of Prisma agent skills | - |
+
+`prisma init` attempts to install `prisma/skills` for detected agents. This is best-effort and does not make project initialization fail. Use `--no-skills` in minimal or controlled environments.
+
+## Examples
+
+### Basic initialization
+
+```bash
+prisma init
+```
+
+Creates a PostgreSQL project setup.
+
+### SQLite project
+
+```bash
+prisma init --datasource-provider sqlite
+```
+
+### MySQL with custom URL
+
+```bash
+prisma init --datasource-provider mysql --url "mysql://user:password@localhost:3306/mydb"
+```
+
+### Prisma Postgres (cloud)
+
+```bash
+prisma init --db
+```
+
+Opens browser for authentication, creates cloud database instance.
+
+### Add an example model
+
+```bash
+prisma init --with-model
+```
+
+Adds a starter model to the generated schema.
+
+### With preview features
+
+```bash
+prisma init --preview-feature relationJoins --preview-feature fullTextSearch
+```
+
+## Generated Schema
+
+```prisma
+generator client {
+  provider = "prisma-client"
+  output   = "../generated/prisma"
+}
+
+datasource db {
+  provider = "postgresql"
+}
+```
+
+## Generated Config (Node.js default)
+
+```typescript
+// prisma.config.ts
+import "dotenv/config";
+import { defineConfig } from 'prisma/config'
+
+export default defineConfig({
+  schema: 'prisma/schema.prisma',
+  migrations: {
+    path: 'prisma/migrations',
+  },
+  datasource: {
+    url: process.env['DATABASE_URL'],
+  },
+})
+```
+
+## Generated Config (Bun)
+
+```typescript
+import { defineConfig, env } from 'prisma/config'
+
+export default defineConfig({
+  schema: 'prisma/schema.prisma',
+  migrations: {
+    path: 'prisma/migrations',
+  },
+  datasource: {
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+## Next Steps After Init
+
+1. Configure `DATABASE_URL` in `.env` (and let `prisma.config.ts` read it)
+2. Define your models in `prisma/schema.prisma`
+3. Run `prisma dev` for local development or connect to remote DB
+4. Run `prisma migrate dev` to create migrations
+5. Run `prisma generate` to generate Prisma Client
+6. Run `prisma db seed` explicitly if you want seed data

.agents/skills/prisma-cli/references/mcp.md

@@ -0,0 +1,39 @@
+# prisma mcp
+
+Starts Prisma's MCP server for AI development tools.
+
+## Command
+
+```bash
+prisma mcp
+```
+
+## What It Does
+
+- Starts a Model Context Protocol (MCP) server for your Prisma project
+- Exposes Prisma schema and database context to compatible AI tools
+- Helps AI assistants understand models, generate queries, and suggest migrations
+
+## Usage
+
+```bash
+prisma mcp
+```
+
+## Typical Use Cases
+
+- Connect Prisma to ChatGPT, Claude, or other MCP-aware tools
+- Give an AI assistant access to your Prisma schema structure
+- Help an agent propose queries, schema updates, and migration steps with project context
+
+## Notes
+
+- Run this from the project that contains your Prisma schema and `prisma.config.ts`
+- The command is separate from Prisma Studio and does not open a browser UI
+- The MCP server exposes `migrate-status`, `migrate-dev`, and Prisma Studio tooling. It does not expose the destructive `migrate-reset` tool; do not claim it is available or try to bypass that safety boundary.
+- For destructive shell commands, follow `agent-safety.md` and obtain explicit user consent.
+
+## References
+
+- [Prisma CLI `mcp` command](https://docs.prisma.io/docs/cli/mcp)
+- [Prisma MCP Server](https://www.prisma.io/docs/ai/tools/chatgpt)

.agents/skills/prisma-cli/references/migrate-deploy.md

@@ -0,0 +1,127 @@
+# prisma migrate deploy
+
+Applies pending migrations in production/staging environments.
+
+## Command
+
+```bash
+prisma migrate deploy
+```
+
+## What It Does
+
+- Applies all pending migrations from `prisma/migrations/`
+- Updates `_prisma_migrations` table
+- Does NOT generate new migrations
+- Does NOT run seed scripts
+- Safe for CI/CD and production
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Custom path to your Prisma schema |
+| `--config` | Custom path to your Prisma config file |
+
+## When to Use
+
+- Production deployments
+- Staging environments  
+- CI/CD pipelines
+- Any non-development environment
+
+## Examples
+
+### Basic deployment
+
+```bash
+prisma migrate deploy
+```
+
+### In CI/CD pipeline
+
+```yaml
+# GitHub Actions example
+- name: Apply migrations
+  run: npx prisma migrate deploy
+  env:
+    DATABASE_URL: ${{ secrets.DATABASE_URL }}
+```
+
+### Docker deployment
+
+```dockerfile
+# Run migrations before starting app
+CMD npx prisma migrate deploy && node dist/index.js
+```
+
+## Comparison with migrate dev
+
+| Feature | migrate dev | migrate deploy |
+|---------|-------------|----------------|
+| Creates migrations | Yes | No |
+| Applies migrations | Yes | Yes |
+| Detects drift | Yes | No |
+| Prompts for input | Yes | No |
+| Uses shadow database | Yes | No |
+| Safe for production | No | Yes |
+| Resets on issues | Prompts | Fails |
+
+## Production Workflow
+
+1. **Development**: Create migrations locally
+   ```bash
+   prisma migrate dev --name add_feature
+   ```
+
+2. **Commit**: Include migration files in version control
+   ```bash
+   git add prisma/migrations
+   git commit -m "Add feature migration"
+   ```
+
+3. **Deploy**: Apply in production
+   ```bash
+   prisma migrate deploy
+   ```
+
+## Error Handling
+
+### Failed migration
+
+If a migration fails, `migrate deploy` exits with error. The failed migration is marked as failed in `_prisma_migrations`.
+
+To fix:
+1. Resolve the issue (fix SQL, database state, etc.)
+2. Mark as resolved: `prisma migrate resolve --applied <migration_name>`
+3. Re-run: `prisma migrate deploy`
+
+### Check status first
+
+```bash
+prisma migrate status
+```
+
+Shows pending and applied migrations before deploying.
+
+## Configuration
+
+Ensure `prisma.config.ts` has the production database URL:
+
+```typescript
+import 'dotenv/config'
+import { defineConfig, env } from 'prisma/config'
+
+export default defineConfig({
+  datasource: {
+    url: env('DATABASE_URL'),
+  },
+})
+```
+
+## Best Practices
+
+1. Always run `migrate status` before `migrate deploy` in CI
+2. Have a rollback plan (backup before migrations)
+3. Test migrations in staging first
+4. Never use `migrate dev` in production

.agents/skills/prisma-cli/references/migrate-dev.md

@@ -0,0 +1,145 @@
+# prisma migrate dev
+
+Creates and applies migrations during development. Requires a shadow database.
+
+## Command
+
+```bash
+prisma migrate dev [options]
+```
+
+## What It Does
+
+1. Runs existing migrations in shadow database to detect drift
+2. Applies any pending migrations
+3. Generates new migration from schema changes
+4. Applies new migration to development database
+5. Updates `_prisma_migrations` table
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--name` / `-n` | Name the migration |
+| `--create-only` | Create a new migration but do not apply it |
+| `--schema` | Custom path to your Prisma schema |
+| `--config` | Custom path to your Prisma config file |
+| `--url` | Override the datasource URL from the Prisma config file |
+
+### Follow-up Commands
+
+- Run `prisma generate` explicitly when you need refreshed client output
+- Run `prisma db seed` explicitly when you need seed data
+
+Run `prisma generate` as an explicit follow-up when you need refreshed generated artifacts. Do not rely on historical CLI help that described generators as part of `migrate dev`.
+
+## Examples
+
+### Create and apply migration
+
+```bash
+prisma migrate dev
+```
+
+Prompts for migration name if schema changed.
+
+### Named migration
+
+```bash
+prisma migrate dev --name add_users_table
+```
+
+### Create without applying
+
+```bash
+prisma migrate dev --create-only
+```
+
+Useful for reviewing migration SQL before applying.
+
+### Full workflow
+
+```bash
+prisma migrate dev --name my_migration
+prisma generate
+prisma db seed
+```
+
+## Migration Files
+
+Created in `prisma/migrations/`:
+
+```
+prisma/migrations/
+├── 20240115120000_add_users_table/
+│   └── migration.sql
+├── 20240116090000_add_posts/
+│   └── migration.sql
+└── migration_lock.toml
+```
+
+## Schema Drift Detection
+
+If `migrate dev` detects drift (manual database changes or edited migrations), it prompts to reset:
+
+```
+Drift detected: Your database schema is not in sync.
+
+Do you want to reset your database? All data will be lost.
+```
+
+## When to Use
+
+- Local development
+- Adding new models/fields
+- Changing relations
+- Creating indexes
+
+## When NOT to Use
+
+- Production deployments (use `migrate deploy`)
+- CI/CD pipelines (use `migrate deploy`)
+- MongoDB (use `db push` instead)
+
+## Common Patterns
+
+### After schema changes
+
+```prisma
+// schema.prisma - Add new field
+model User {
+  id        Int      @id @default(autoincrement())
+  email     String   @unique
+  name      String?
+  createdAt DateTime @default(now())  // New field
+}
+```
+
+```bash
+prisma migrate dev --name add_created_at
+```
+
+### Handling data loss warnings
+
+When a migration would cause data loss:
+
+```bash
+prisma migrate dev --name remove_field
+# Warning: You are about to delete data...
+# Accept with: --accept-data-loss
+```
+
+## Shadow Database
+
+`migrate dev` requires a shadow database for drift detection. Configure in `prisma.config.ts`:
+
+```typescript
+export default defineConfig({
+  datasource: {
+    url: env('DATABASE_URL'),
+    shadowDatabaseUrl: env('SHADOW_DATABASE_URL'),
+  },
+})
+```
+
+For local Prisma Postgres (`prisma dev`), shadow database is handled automatically.

.agents/skills/prisma-cli/references/migrate-diff.md

@@ -0,0 +1,89 @@
+# prisma migrate diff
+
+Compares database schemas and generates diffs (SQL or summary).
+
+## Command
+
+```bash
+prisma migrate diff [options]
+```
+
+## What It Does
+
+- Compares two sources (`--from-...` and `--to-...`)
+- Sources can be:
+    - Empty (`empty`)
+    - Schema file (`schema`)
+    - Migrations directory (`migrations`)
+    - Database URL (`url`) or Configured Datasource (`config-datasource`)
+- Outputs the difference:
+    - Human-readable summary (default)
+    - SQL script (`--script`)
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--script` | Render SQL script to stdout |
+| `--exit-code` | Exit 2 if changes detected, 0 if empty, 1 if error |
+| `--config` | Custom path to your Prisma config file |
+
+### Sources (Must provide one `from` and one `to`)
+
+- `--from-empty`, `--to-empty`
+- `--from-schema <path>`, `--to-schema <path>`
+- `--from-migrations <path>`, `--to-migrations <path>`
+- `--from-url <url>`, `--to-url <url>`
+- `--from-config-datasource`, `--to-config-datasource` (uses `prisma.config.ts`)
+
+## Examples
+
+### Generate SQL for a schema change
+
+Compare current production DB to your local schema:
+
+```bash
+prisma migrate diff \
+  --from-url "$PROD_DB_URL" \
+  --to-schema ./prisma/schema.prisma \
+  --script
+```
+
+### Review pending migrations
+
+Compare database state to migrations directory:
+
+```bash
+prisma migrate diff \
+  --from-config-datasource \
+  --to-migrations ./prisma/migrations
+```
+
+### Create baseline migration
+
+Compare empty state to current schema:
+
+```bash
+prisma migrate diff \
+  --from-empty \
+  --to-schema ./prisma/schema.prisma \
+  --script > prisma/migrations/0_init/migration.sql
+```
+
+### Check for drift (CI)
+
+Check if database matches schema:
+
+```bash
+prisma migrate diff \
+  --from-config-datasource \
+  --to-schema ./prisma/schema.prisma \
+  --exit-code
+```
+
+## Use Cases
+
+- **Forward-generating migrations**: Creating SQL without `migrate dev`.
+- **Drift detection**: Checking if DB is in sync.
+- **Baselining**: Creating initial migration from existing DB.
+- **Debugging**: Understanding what `migrate dev` would do.

.agents/skills/prisma-cli/references/migrate-reset.md

@@ -0,0 +1,80 @@
+# prisma migrate reset
+
+Resets your database and re-applies all migrations.
+
+## Command
+
+```bash
+prisma migrate reset [options]
+```
+
+## What It Does
+
+1. **Drops** the database (if possible) or deletes all data/tables
+2. **Re-creates** the database
+3. **Applies** all migrations from `prisma/migrations/`
+4. Stops there - run seed and generate explicitly if needed
+
+**Warning: All data will be lost.**
+
+When Prisma detects an AI agent, this command is blocked until the user gives explicit consent. Follow `agent-safety.md`; `--force` skips the ordinary prompt but does not constitute user consent for an agent.
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--force` / `-f` | Skip confirmation prompt |
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Examples
+
+### Basic reset
+
+```bash
+prisma migrate reset
+```
+
+Prompts for confirmation in interactive terminals.
+
+### Force reset (CI/Automation)
+
+```bash
+prisma migrate reset --force
+```
+
+### With custom schema
+
+```bash
+prisma migrate reset --schema=./custom/schema.prisma
+```
+
+## When to Use
+
+- **Development**: When you want a fresh start
+- **Testing**: Resetting test database before suites
+- **Drift Recovery**: When the database is out of sync and you can't migrate
+
+## Follow-up Steps
+
+Run `prisma generate` and `prisma db seed` explicitly when you need refreshed client output or seed data after a reset.
+
+## Configuration
+
+Configure the seed script in `prisma.config.ts`, then run it explicitly after reset:
+
+```typescript
+export default defineConfig({
+  migrations: {
+    seed: 'tsx prisma/seed.ts',
+  },
+})
+```
+
+Typical workflow:
+
+```bash
+prisma migrate reset --force
+prisma generate
+prisma db seed
+```

.agents/skills/prisma-cli/references/migrate-resolve.md

@@ -0,0 +1,57 @@
+# prisma migrate resolve
+
+Resolves issues with database migrations, such as failed migrations or baselining.
+
+## Command
+
+```bash
+prisma migrate resolve [options]
+```
+
+## What It Does
+
+Updates the `_prisma_migrations` table to manually change the state of a migration. This is a recovery tool.
+
+## Options
+
+You must provide exactly one of `--applied` or `--rolled-back`.
+
+| Option | Description |
+|--------|-------------|
+| `--applied <name>` | Mark a migration as **applied** (success) |
+| `--rolled-back <name>` | Mark a migration as **rolled back** (ignored/failed) |
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Examples
+
+### Mark as Applied (Baselining)
+
+If you have existing tables and want to initialize migrations without running the SQL:
+
+```bash
+prisma migrate resolve --applied 20240101000000_initial_migration
+```
+
+This tells Prisma "Assume this migration has already run".
+
+### Mark as Rolled Back (Fixing Failures)
+
+If a migration failed (e.g., syntax error) and you fixed the SQL or want to retry:
+
+```bash
+prisma migrate resolve --rolled-back 20240115120000_failed_migration
+```
+
+This tells Prisma "Forget this migration run, let me try applying it again".
+
+## Use Cases
+
+1. **Baselining**: Adopting Prisma Migrate on an existing production database.
+2. **Failed Migrations**: Recovering from a failed `migrate deploy` in production.
+3. **Hotfixes**: reconciling manual database changes (rare).
+
+## References
+
+- [Baselining](https://www.prisma.io/docs/guides/database/developing-with-prisma-migrate/baselining)
+- [Troubleshooting](https://www.prisma.io/docs/guides/database/production-troubleshooting)

.agents/skills/prisma-cli/references/migrate-status.md

@@ -0,0 +1,65 @@
+# prisma migrate status
+
+Checks the status of your database migrations.
+
+## Command
+
+```bash
+prisma migrate status [options]
+```
+
+## What It Does
+
+- Connects to the database
+- Checks the `_prisma_migrations` table
+- Compares applied migrations with local migration files
+- Reports:
+    - **Status**: Database is up-to-date or behind
+    - **Unapplied migrations**: Count of pending migrations
+    - **Missing migrations**: Migrations present in DB but missing locally
+    - **Failed migrations**: Any migrations that failed to apply
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Examples
+
+### Check status
+
+```bash
+prisma migrate status
+```
+
+Output example (Up to date):
+```
+Database schema is up to date!
+```
+
+Output example (Pending):
+```
+Following migration have not yet been applied:
+  20240115120000_add_user
+
+To apply migrations in development, run:
+  prisma migrate dev
+
+To apply migrations in production, run:
+  prisma migrate deploy
+```
+
+## When to Use
+
+- **Debugging**: Why is `migrate dev` complaining about drift?
+- **CI/CD**: Verify database state before deploying
+- **Production**: Check if migrations are needed (`migrate deploy`) or if a deployment failed
+
+## Exit Codes
+
+- `0`: Success (may have pending migrations, but command ran successfully)
+- `1`: Error
+
+To check for pending migrations programmatically, you might need to parse the output or use `migrate diff` with exit code flags.

.agents/skills/prisma-cli/references/studio.md

@@ -0,0 +1,137 @@
+# prisma studio
+
+Opens a visual database browser for viewing and editing data.
+
+## Command
+
+```bash
+prisma studio [options]
+```
+
+## What It Does
+
+- Starts a web-based database GUI
+- View all your models and records
+- Create, update, and delete records
+- Filter and sort data
+- Navigate relations
+
+## Options
+
+| Option | Description | Default |
+|--------|-------------|---------|
+| `--port` / `-p` | Port to start Studio on | `5555` |
+| `--browser` / `-b` | Browser to open Studio in | System default |
+| `--config` | Custom path to your Prisma config file | - |
+| `--url` | Database connection string (overrides the one in your Prisma config) | - |
+
+## Examples
+
+### Open Studio
+
+```bash
+prisma studio
+```
+
+Opens at http://localhost:5555
+
+### Custom port
+
+```bash
+prisma studio --port 3000
+```
+
+### Specific browser
+
+```bash
+prisma studio --browser firefox
+```
+
+### Don't open browser
+
+```bash
+BROWSER=none prisma studio
+```
+
+Useful for remote servers.
+
+## Features
+
+### View Records
+
+- See all records in table format
+- Pagination for large datasets
+- Column sorting
+
+### Filter Data
+
+- Filter by any field
+- Multiple conditions
+- Relation filtering
+
+### Edit Records
+
+- Click to edit inline
+- Add new records
+- Delete records (with confirmation)
+
+### Navigate Relations
+
+- Click relations to view related records
+- See counts of related items
+- Follow relation links
+
+## Recent Studio Capabilities
+
+Recent Prisma Studio releases added richer editor workflows:
+
+- multi-cell selection and editing
+- full-table search and more intuitive filtering
+- command palette shortcuts
+- dark mode
+- copy selections as Markdown
+- back-relation navigation
+- SQL workflows including raw SQL queries
+
+Some recent builds also expose AI-assisted SQL authoring. Treat these as interactive Studio features rather than a replacement for checked-in migrations or application queries.
+
+## Use Cases
+
+- **Development**: Quick data inspection
+- **Debugging**: Check data state
+- **Testing**: Verify seed data
+- **Demo**: Show data to stakeholders
+
+## Limitations
+
+- Development tool only
+- Not for production use
+- Limited to configured database
+- Prisma Studio in Prisma 7 currently targets PostgreSQL, MySQL, and SQLite first
+- For reproducible application logic, prefer Prisma Client and checked-in SQL scripts
+
+## Common Workflow
+
+1. Run migrations:
+   ```bash
+   prisma migrate dev
+   ```
+
+2. Seed data:
+   ```bash
+   prisma db seed
+   ```
+
+3. Open Studio to verify:
+   ```bash
+   prisma studio
+   ```
+
+4. Make manual edits if needed
+
+## Security Note
+
+Studio provides direct database access. Only run on:
+- Local development machines
+- Secure internal networks
+- Never expose publicly

.agents/skills/prisma-cli/references/validate.md

@@ -0,0 +1,53 @@
+# prisma validate
+
+Validates your Prisma schema file.
+
+## Command
+
+```bash
+prisma validate [options]
+```
+
+## What It Does
+
+- Parses the `schema.prisma` file
+- Checks for syntax errors
+- Validates model definitions, relations, and types
+- Reports any errors or warnings without generating code
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--schema` | Path to schema file |
+| `--config` | Custom path to your Prisma config file |
+
+## Examples
+
+### Validate default schema
+
+```bash
+prisma validate
+```
+
+### Validate specific schema
+
+```bash
+prisma validate --schema=./custom/schema.prisma
+```
+
+### Use in CI
+
+Run `validate` in your CI pipeline to catch schema errors early:
+
+```yaml
+- name: Validate Schema
+  run: npx prisma validate
+```
+
+## Common Errors
+
+- Missing `@relation` fields
+- Invalid types
+- Duplicate model names
+- Syntax errors (missing braces, etc.)