/csv-to-sql <csv-path>
/csv-to-sql Inbox/data.csv
/csv-to-sql Inbox/rfi.csv --start-row 14 --fts
/csv-to-sql Inbox/survey.csv --db-name survey-2026.db --table responses

node scripts/csv-to-sqlite.js "<csv-path>" --start-row <n> --fts --verbose

# Show table structure
sqlite3 .data/<db-name>.db ".schema"

# Preview data
sqlite3 .data/<db-name>.db -markdown -header "SELECT * FROM data LIMIT 5;"

# Count rows
sqlite3 .data/<db-name>.db "SELECT COUNT(*) FROM data;"

## CSV to SQLite Complete

**Database**: `.data/<db-name>.db`
**Table**: `<table-name>`

**Statistics**:

- Rows: <count>
- Columns: <count>
- Indexes: <list>
- FTS enabled: Yes/No

**Quick Queries**:

```bash
# List all records
sqlite3 ".data/<db>.db" -markdown -header "SELECT * FROM <table>;"

# Aggregate by category
sqlite3 ".data/<db>.db" -markdown -header "SELECT category, COUNT(*) FROM <table> GROUP BY category;"

# Search text (if FTS enabled)
sqlite3 ".data/<db>.db" "SELECT * FROM <table>_fts WHERE <table>_fts MATCH 'keyword';"
```

## Common Query Patterns

### For RFI/Survey Data

```sql
-- Average score by category
SELECT category, ROUND(AVG(score), 2) as avg_score, COUNT(*) as questions
FROM data GROUP BY category ORDER BY avg_score DESC;

-- Questions with lowest scores
SELECT id, question, category, score
FROM data WHERE score < 6 ORDER BY score ASC;

-- Score distribution
SELECT score, COUNT(*) as count
FROM data GROUP BY score ORDER BY score;

-- Search responses for keyword
SELECT id, question, response
FROM data_fts WHERE data_fts MATCH 'integration pattern';

-- Distinct values in a column
SELECT DISTINCT category FROM data;

-- Records matching criteria
SELECT * FROM data WHERE status = 'active' AND priority = 'high';

-- Export to JSON
-- Use: sqlite3 .data/db.db -json "SELECT * FROM data;"

YourVault/
├── .data/
│   ├── vendor-rfi-scoring.db
│   ├── vendor-evaluation.db
│   └── survey-results.db
└── Inbox/
    └── source-data.csv

-- Search in any text column
SELECT * FROM data_fts WHERE data_fts MATCH 'middleware integration';

-- Search with ranking
SELECT *, rank FROM data_fts WHERE data_fts MATCH 'API design' ORDER BY rank;

-- Phrase search
SELECT * FROM data_fts WHERE data_fts MATCH '"backward compatibility"';

# From Claude Code, run queries via Bash tool
sqlite3 .data/vendor-rfi-scoring.db -markdown -header "
SELECT category, COUNT(*) as questions, ROUND(AVG(score),1) as avg
FROM data
GROUP BY category
ORDER BY avg DESC;"

User: /csv-to-sql Inbox/vendor-rfi-responses.csv

Claude:
1. Runs: node scripts/csv-to-sqlite.js "..." --dry-run --verbose
2. Identifies start-row needed (14)
3. Runs: node scripts/csv-to-sqlite.js "..." --start-row 14 --fts --verbose
4. Verifies: sqlite3 .data/... -markdown -header "SELECT * FROM data LIMIT 3;"
5. Reports summary with example queries