Validate with Database
Connect to live PostgreSQL database to validate schema assumptions, compare pg_dump vs pgschema output, and query system catalogs interactively
Best use case
Validate with Database is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Connect to live PostgreSQL database to validate schema assumptions, compare pg_dump vs pgschema output, and query system catalogs interactively
Teams using Validate with Database should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/validate-with-database/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How Validate with Database Compares
| Feature / Agent | Validate with Database | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Connect to live PostgreSQL database to validate schema assumptions, compare pg_dump vs pgschema output, and query system catalogs interactively
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Validate with Database
Use this skill to connect to the test PostgreSQL database, validate assumptions about schema behavior, and cross-validate between pg_dump and pgschema implementations.
## When to Use This Skill
Invoke this skill when:
- Validating how PostgreSQL actually stores or represents schema objects
- Comparing pg_dump output with pgschema output
- Testing a new feature implementation against real database
- Debugging schema introspection issues
- Verifying system catalog query results
- Understanding how PostgreSQL formats specific DDL
- Checking version-specific behavior (PostgreSQL 14-17)
- Validating migration plans before implementing new features
## Database Connection Information
Connection details are stored in `.env` file at project root:
```
PGHOST=localhost
PGDATABASE=employee
PGUSER=postgres
PGPASSWORD=testpwd1
```
**Default connection**:
- Host: `localhost`
- Port: `5432` (default)
- Database: `employee`
- User: `postgres`
- Password: `testpwd1`
## Connection Methods
### Method 1: Using psql (Interactive Queries)
**Basic connection**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee
```
**One-off query**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee -c "SELECT version();"
```
**Execute multi-line query**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
SELECT
t.tgname,
CASE
WHEN t.tgqual IS NOT NULL
THEN pg_get_expr(t.tgqual, t.tgrelid, false)
ELSE 'NO WHEN CLAUSE'
END as when_clause
FROM pg_catalog.pg_trigger t
JOIN pg_catalog.pg_class c ON t.tgrelid = c.oid
WHERE c.relname = 'test_table'
ORDER BY t.tgname;
"
```
### Method 2: Using pg_dump (Schema Export)
**Dump entire database schema**:
```bash
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --schema=public
```
**Dump specific table**:
```bash
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --table=employees
```
**Dump only specific object types**:
```bash
# Only triggers
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --schema=public | grep -A 20 "CREATE TRIGGER"
# Only indexes
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --schema=public | grep -A 10 "CREATE INDEX"
```
### Method 3: Using pgschema (Project Tool)
**Dump with pgschema**:
```bash
./pgschema dump --host localhost --port 5432 --db employee --user postgres --schema public
```
**Or using environment variables** (from .env):
```bash
# .env is automatically loaded by pgschema
./pgschema dump --schema public
```
**Dump to file**:
```bash
./pgschema dump --schema public -o /tmp/schema_dump.sql
```
### Method 4: Database Setup for Testing
**Create a test database**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "DROP DATABASE IF EXISTS test_validation;"
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "CREATE DATABASE test_validation;"
```
**Create test schema objects**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d test_validation -c "
CREATE TABLE test_table (
id SERIAL PRIMARY KEY,
name TEXT NOT NULL,
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
CREATE TRIGGER test_trigger
BEFORE INSERT ON test_table
FOR EACH ROW
WHEN (NEW.name IS NOT NULL)
EXECUTE FUNCTION my_trigger_func();
"
```
## Common Validation Workflows
### Workflow 1: Compare pg_dump vs pgschema Output
**Purpose**: Verify pgschema produces comparable output to pg_dump
**Steps**:
1. **Dump with pg_dump**:
```bash
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --schema=public > /tmp/pg_dump_output.sql
```
2. **Dump with pgschema**:
```bash
./pgschema dump --schema public -o /tmp/pgschema_output.sql
```
3. **Compare outputs**:
```bash
# Side-by-side comparison
diff -u /tmp/pg_dump_output.sql /tmp/pgschema_output.sql
# Or use a better diff tool
code --diff /tmp/pg_dump_output.sql /tmp/pgschema_output.sql
```
4. **Analyze differences**:
- Formatting differences (expected)
- Missing objects (bugs to fix)
- Different DDL structure (may need investigation)
- Comments handling
- Ordering differences
### Workflow 2: Validate System Catalog Queries
**Purpose**: Test system catalog queries return expected data
**Steps**:
1. **Create test object**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
CREATE TABLE test_triggers (
id INTEGER PRIMARY KEY,
data TEXT
);
CREATE OR REPLACE FUNCTION trigger_func() RETURNS TRIGGER AS \$\$
BEGIN
RETURN NEW;
END;
\$\$ LANGUAGE plpgsql;
CREATE TRIGGER test_when_trigger
BEFORE INSERT ON test_triggers
FOR EACH ROW
WHEN (NEW.data <> '')
EXECUTE FUNCTION trigger_func();
"
```
2. **Query system catalogs**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
SELECT
t.tgname,
t.tgtype,
CASE
WHEN t.tgqual IS NOT NULL
THEN pg_get_expr(t.tgqual, t.tgrelid, false)
ELSE NULL
END as when_clause,
pg_get_triggerdef(t.oid) as full_definition
FROM pg_catalog.pg_trigger t
JOIN pg_catalog.pg_class c ON t.tgrelid = c.oid
WHERE c.relname = 'test_triggers'
AND t.tgisinternal = false;
"
```
3. **Verify pgschema extracts same data**:
```bash
./pgschema dump --schema public | grep -A 20 "test_when_trigger"
```
4. **Cleanup**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
DROP TRIGGER IF EXISTS test_when_trigger ON test_triggers;
DROP TABLE IF EXISTS test_triggers;
DROP FUNCTION IF EXISTS trigger_func();
"
```
### Workflow 3: Test Plan/Apply Workflow
**Purpose**: Validate pgschema plan and apply work correctly
**Steps**:
1. **Create initial schema**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
DROP SCHEMA IF EXISTS test_workflow CASCADE;
CREATE SCHEMA test_workflow;
SET search_path TO test_workflow;
CREATE TABLE users (
id SERIAL PRIMARY KEY,
email TEXT NOT NULL UNIQUE
);
"
```
2. **Dump current state**:
```bash
./pgschema dump --schema test_workflow -o /tmp/current_schema.sql
```
3. **Modify schema file** (edit /tmp/current_schema.sql):
```sql
-- Add a new column
CREATE TABLE users (
id SERIAL PRIMARY KEY,
email TEXT NOT NULL UNIQUE,
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP -- NEW
);
```
4. **Generate plan**:
```bash
./pgschema plan --schema test_workflow --file /tmp/current_schema.sql
```
5. **Review migration DDL** - should show:
```sql
ALTER TABLE users ADD COLUMN created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP;
```
6. **Apply migration**:
```bash
./pgschema apply --schema test_workflow --file /tmp/current_schema.sql --auto-approve
```
7. **Verify result**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "\d test_workflow.users"
```
8. **Cleanup**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "DROP SCHEMA IF EXISTS test_workflow CASCADE;"
```
### Workflow 4: Validate Specific DDL Formatting
**Purpose**: Understand how PostgreSQL formats specific constructs
**Steps**:
1. **Create object with specific feature**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
CREATE TABLE test_pk_order (
b INTEGER,
a INTEGER,
c INTEGER,
PRIMARY KEY (a, b) -- Note: different order than column definition
);
"
```
2. **Check how PostgreSQL stores it**:
```bash
# Use \d+ to see structure
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "\d+ test_pk_order"
```
3. **See pg_dump format**:
```bash
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d postgres --schema-only --table=test_pk_order
```
4. **Query system catalogs directly**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
SELECT
c.relname as table_name,
con.conname as constraint_name,
pg_get_constraintdef(con.oid) as constraint_def
FROM pg_constraint con
JOIN pg_class c ON con.conrelid = c.oid
WHERE c.relname = 'test_pk_order';
"
```
5. **Compare with pgschema**:
```bash
./pgschema dump --schema public | grep -A 10 "test_pk_order"
```
### Workflow 5: Cross-Version Testing
**Purpose**: Validate behavior across PostgreSQL versions 14-17
**Steps**:
1. **Check current version**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "SELECT version();"
```
2. **Run version-specific integration tests**:
```bash
# Test against specific version
PGSCHEMA_POSTGRES_VERSION=14 go test -v ./cmd/dump -run TestDumpCommand_Employee
PGSCHEMA_POSTGRES_VERSION=17 go test -v ./cmd/dump -run TestDumpCommand_Employee
```
3. **Check for version-specific features**:
```bash
# PostgreSQL 15+ feature: UNIQUE NULLS NOT DISTINCT
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d postgres -c "
SELECT version();
CREATE TABLE test_nulls (
id INTEGER,
email TEXT UNIQUE NULLS NOT DISTINCT
);
"
```
## Useful System Catalog Queries
### Inspect Tables and Columns
```sql
-- All tables in schema
SELECT schemaname, tablename
FROM pg_tables
WHERE schemaname = 'public';
-- Columns with types
SELECT
a.attname as column_name,
pg_catalog.format_type(a.atttypid, a.atttypmod) as data_type,
a.attnotnull as not_null,
pg_get_expr(ad.adbin, ad.adrelid) as default_value,
a.attgenerated as generated
FROM pg_attribute a
LEFT JOIN pg_attrdef ad ON (a.attrelid = ad.adrelid AND a.attnum = ad.adnum)
WHERE a.attrelid = 'public.employees'::regclass
AND a.attnum > 0
AND NOT a.attisdropped
ORDER BY a.attnum;
```
### Inspect Constraints
```sql
-- All constraints on a table
SELECT
con.conname as constraint_name,
con.contype as constraint_type,
pg_get_constraintdef(con.oid) as definition
FROM pg_constraint con
WHERE con.conrelid = 'public.employees'::regclass;
```
### Inspect Indexes
```sql
-- All indexes on a table
SELECT
i.relname as index_name,
am.amname as index_type,
pg_get_indexdef(idx.indexrelid) as definition,
CASE
WHEN idx.indpred IS NOT NULL
THEN pg_get_expr(idx.indpred, idx.indrelid, true)
ELSE NULL
END as where_clause
FROM pg_index idx
JOIN pg_class i ON i.oid = idx.indexrelid
JOIN pg_class t ON t.oid = idx.indrelid
JOIN pg_am am ON i.relam = am.oid
WHERE t.relname = 'employees'
AND t.relnamespace = 'public'::regnamespace;
```
### Inspect Triggers
```sql
-- All triggers on a table
SELECT
t.tgname as trigger_name,
t.tgenabled as enabled,
CASE t.tgtype::integer & 66
WHEN 2 THEN 'BEFORE'
WHEN 64 THEN 'INSTEAD OF'
ELSE 'AFTER'
END as timing,
pg_get_triggerdef(t.oid) as full_definition,
CASE
WHEN t.tgqual IS NOT NULL
THEN pg_get_expr(t.tgqual, t.tgrelid, false)
ELSE NULL
END as when_condition
FROM pg_trigger t
JOIN pg_class c ON t.tgrelid = c.oid
WHERE c.relname = 'employees'
AND c.relnamespace = 'public'::regnamespace
AND NOT t.tgisinternal;
```
### Inspect Views and Materialized Views
```sql
-- Views
SELECT
schemaname,
viewname,
definition
FROM pg_views
WHERE schemaname = 'public';
-- Materialized views
SELECT
schemaname,
matviewname,
definition
FROM pg_matviews
WHERE schemaname = 'public';
```
### Inspect Comments
```sql
-- Comments on tables
SELECT
c.relname as table_name,
d.description as comment
FROM pg_class c
JOIN pg_namespace n ON n.oid = c.relnamespace
LEFT JOIN pg_description d ON d.objoid = c.oid AND d.objsubid = 0
WHERE n.nspname = 'public'
AND c.relkind = 'r'
AND d.description IS NOT NULL;
-- Comments on columns
SELECT
c.relname as table_name,
a.attname as column_name,
d.description as comment
FROM pg_class c
JOIN pg_namespace n ON n.oid = c.relnamespace
JOIN pg_attribute a ON a.attrelid = c.oid
LEFT JOIN pg_description d ON d.objoid = c.oid AND d.objsubid = a.attnum
WHERE n.nspname = 'public'
AND c.relkind = 'r'
AND a.attnum > 0
AND NOT a.attisdropped
AND d.description IS NOT NULL;
```
## Troubleshooting Common Issues
### Issue: Connection Refused
```bash
# Check if PostgreSQL is running
pg_isready -h localhost -p 5432
# Check if port is open
nc -zv localhost 5432
# Check PostgreSQL logs
tail -f /usr/local/var/postgresql@14/server.log # macOS Homebrew
```
### Issue: Authentication Failed
```bash
# Verify credentials
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "SELECT current_user;"
# Check pg_hba.conf settings
cat /usr/local/var/postgresql@14/pg_hba.conf
```
### Issue: Database Doesn't Exist
```bash
# List all databases
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "\l"
# Create if missing
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "CREATE DATABASE employee;"
```
### Issue: Schema Not Found
```bash
# List all schemas
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee -c "\dn"
# Create if missing
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee -c "CREATE SCHEMA public;"
```
## Test Data Setup
### Load Sample Database
The project includes test data in `testdata/dump/`:
```bash
# Employee database
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "DROP DATABASE IF EXISTS employee;"
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "CREATE DATABASE employee;"
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee < testdata/dump/employee/employee.sql
# Sakila database (if available)
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "DROP DATABASE IF EXISTS sakila;"
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "CREATE DATABASE sakila;"
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d sakila < testdata/dump/sakila/sakila.sql
```
### Create Minimal Test Database
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres <<EOF
DROP DATABASE IF EXISTS test_minimal;
CREATE DATABASE test_minimal;
\c test_minimal
CREATE SCHEMA app;
CREATE TABLE app.users (
id SERIAL PRIMARY KEY,
email TEXT NOT NULL UNIQUE,
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
CREATE TABLE app.posts (
id SERIAL PRIMARY KEY,
user_id INTEGER NOT NULL REFERENCES app.users(id) ON DELETE CASCADE,
title TEXT NOT NULL,
content TEXT,
published_at TIMESTAMP
);
CREATE INDEX idx_posts_user ON app.posts(user_id);
CREATE INDEX idx_posts_published ON app.posts(published_at) WHERE published_at IS NOT NULL;
COMMENT ON TABLE app.users IS 'Application users';
COMMENT ON COLUMN app.users.email IS 'User email address (must be unique)';
EOF
```
## Quick Reference Commands
**Connect to database**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee
```
**Run query**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee -c "SELECT * FROM employees LIMIT 5;"
```
**Describe table**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -d employee -c "\d+ employees"
```
**pg_dump schema only**:
```bash
PGPASSWORD='testpwd1' pg_dump -h localhost -p 5432 -U postgres -d employee --schema-only --schema=public
```
**pgschema dump**:
```bash
./pgschema dump --schema public
```
**pgschema plan**:
```bash
./pgschema plan --schema public --file schema.sql
```
**Drop and recreate test database**:
```bash
PGPASSWORD='testpwd1' psql -h localhost -p 5432 -U postgres -c "DROP DATABASE IF EXISTS test_db; CREATE DATABASE test_db;"
```
## Validation Checklist
When validating implementation:
- [ ] Test database is running and accessible
- [ ] Connection credentials from .env work
- [ ] pg_dump produces expected output
- [ ] pgschema produces comparable output
- [ ] System catalog queries return expected data
- [ ] DDL formatting matches PostgreSQL conventions
- [ ] Plan generates correct migration DDL
- [ ] Apply successfully executes migration
- [ ] Final state matches expected schema
- [ ] Tested across PostgreSQL versions (if version-specific)
- [ ] Test database cleaned up after validationRelated Skills
docker-database
Configure database containers with security, persistence, and health checks
Database Sync
Automate database synchronization, replication, migration, and cross-platform data integration
database-skill
Design and manage relational databases including table creation, migrations, and schema design. Use for database modeling and maintenance.
database-architect
Database design and optimization specialist. Schema design, query optimization, indexing strategies, data modeling, and migration planning for relational and NoSQL databases.
arch-database
DB architecture: relational vs document vs graph vs vector, schema design, indexing, replication, sharding
acsets-algebraic-databases
ACSets (Attributed C-Sets): Algebraic databases as in-memory data structures. Category-theoretic formalism for relational databases generalizing graphs and data frames.
vercel-kv-database-rules
Defines how to interact with Vercel's KV database for storing and retrieving session and application data.
sqlmap-database-pentesting
This skill should be used when the user asks to "automate SQL injection testing," "enumerate database structure," "extract database credentials using sqlmap," "dump tables and columns...
relational-database-mcp-cloudbase
This is the required documentation for agents operating on the CloudBase Relational Database. It lists the only four supported tools for running SQL and managing security rules. Read the full content to understand why you must NOT use standard Application SDKs and how to safely execute INSERT, UPDATE, or DELETE operations without corrupting production data.
railway-database
Add official Railway database services (Postgres, Redis, MySQL, MongoDB). Use when user wants to add a database, says "add postgres", "add redis", "add database", "connect to database", or "wire up the database". For other templates (Ghost, Strapi, n8n), use the railway-templates skill.
prisma-database-setup
Guides for configuring Prisma with different database providers (PostgreSQL, MySQL, SQLite, MongoDB, etc.). Use when setting up a new project, changing databases, or troubleshooting connection issues. Triggers on "configure postgres", "connect to mysql", "setup mongodb", "sqlite setup".
playerhoods-database
Authoritative database governance, schema, and RPC invariant model for playerhoods.com (docs v1.4 aligned). Enforces club identity system and RPC-only write posture.