Phase 1b: Rename all ~94 commands across 12 plugins to /<noun> <action> sub-command pattern. Git-flow consolidated from 8→5 commands (commit variants absorbed into --push/--merge/--sync flags). Dispatch files, name: frontmatter, and cross-reference updates for all plugins. Phase 2: Design documents for 8 new plugins in docs/designs/. Phase 3: Scaffold 8 new plugins — saas-api-platform, saas-db-migrate, saas-react-platform, saas-test-pilot, data-seed, ops-release-manager, ops-deploy-pipeline, debug-mcp. Each with plugin.json, commands, agents, skills, README, and claude-md-integration. Marketplace grows from 12→20. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2.2 KiB
2.2 KiB
data-platform Plugin - CLAUDE.md Integration
Add this section to your project's CLAUDE.md to enable data-platform plugin features.
Suggested CLAUDE.md Section
## Data Platform Integration
This project uses the data-platform plugin for data engineering workflows.
### Configuration
**PostgreSQL**: Credentials in `~/.config/claude/postgres.env`
**dbt**: Project path auto-detected from `dbt_project.yml`
### Available Commands
| Command | Purpose |
|---------|---------|
| `/data ingest` | Load data from files or database |
| `/data profile` | Generate statistical profile |
| `/data schema` | Show schema information |
| `/data explain` | Explain dbt model |
| `/data lineage` | Show data lineage |
| `/data run` | Execute dbt models |
### data_ref Convention
DataFrames are stored with references. Use meaningful names:
- `raw_*` for source data
- `stg_*` for staged/cleaned data
- `dim_*` for dimension tables
- `fct_*` for fact tables
- `rpt_*` for reports
### dbt Workflow
1. Always validate before running: `/data run` includes automatic `dbt_parse`
2. For dbt 1.9+, check for deprecated syntax before commits
3. Use `/data lineage` to understand impact of changes
### Database Access
PostgreSQL tools require POSTGRES_URL configuration:
- Read-only queries: `pg_query`
- Write operations: `pg_execute`
- Schema exploration: `pg_tables`, `pg_columns`
PostGIS spatial data:
- List spatial tables: `st_tables`
- Check geometry: `st_geometry_type`, `st_srid`, `st_extent`
Environment Variables
Add to project .env if needed:
# dbt configuration
DBT_PROJECT_DIR=./transform
DBT_PROFILES_DIR=~/.dbt
# Memory limits
DATA_PLATFORM_MAX_ROWS=100000
Typical Workflows
Data Exploration
/data ingest data/raw_customers.csv
/data profile raw_customers
/data schema
ETL Development
/data schema orders # Understand source
/data explain stg_orders # Understand transformation
/data run stg_orders # Test the model
/data lineage fct_orders # Check downstream impact
Database Analysis
/data schema # List all tables
pg_columns orders # Detailed schema
st_tables # Find spatial data