CQLAI
CQLAI is a modern, AI-powered CQL shell built by AxonOps as an alternative to cqlsh. It provides a rich terminal interface, zero dependencies, and optional AI-powered query generation from natural language.
Why CQLAI for Development?
| cqlsh Limitation |
CQLAI Solution |
| Requires Python installation |
Single static binary, no dependencies |
| Slow startup (2-3 seconds) |
Fast startup (~100ms) |
| Basic terminal interface |
Rich TUI with mouse support |
| Limited output formats |
Table, JSON, CSV, Parquet export |
| No AI assistance |
Natural language to CQL conversion |
| Can OOM on large results |
Memory-bounded virtualized display |
Key Features
- Zero Dependencies — Single binary for Linux, macOS, Windows
- Rich Terminal UI — Full-screen mode, vim navigation, mouse support
- AI Query Generation — Convert natural language to CQL (optional)
- Apache Parquet Support — Native import/export for analytics integration
- Smart Auto-Completion — Context-aware completion for tables, columns, keywords
Quick Example
# Connect to Cassandra
cqlai --host cassandra.example.com -u myuser -k my_keyspace
# Run standard CQL
cqlai> SELECT * FROM users LIMIT 10;
# Use AI to generate queries
cqlai> .ai find all orders placed today with total over 100
# Export to Parquet
cqlai> COPY users TO 'users.parquet' WITH COMPRESSION='ZSTD';
Full Documentation
For complete documentation including installation, configuration, and all features, see the CQLAI Tools Reference.
| Section |
Description |
| Installation |
APT, YUM, binary download, Docker, build from source |
| Getting Started |
Connection options, authentication, SSL/TLS |
| Configuration |
Config files, environment variables, options |
| AI Features |
AI providers, natural language queries |
| Data Import/Export |
CSV, Parquet, COPY commands |
| Commands |
All supported commands |
| Troubleshooting |
Common issues and solutions |
Resources