
TESH-Query
Passionate about democratizing data access and building intelligent developer tools.
Free440 impressions#1670 of its week
TESH-Query is a CLI and Python SDK that converts natural language questions into safe SQL queries and returns results.
- for
- Developers and data teams who need SQL insights without writing SQL code.
- pricing
- open source
- license
- AGPL-3.0
- v3.0.0.dev118 days ago
- v2.1.12 months ago
- v2.1.02 months ago
Key features
- Natural Language Querying — Ask questions in plain English and get SQL plus results instantly.
- Multi-LLM Support — Choose Google Gemini or Azure OpenAI as the underlying model.
- Broad Database Compatibility — Works with PostgreSQL, MySQL, SQLite, MSSQL, MariaDB, Oracle via SQLAlchemy.
- Safe Execution — Generated SQL is AST-validated and destructive commands are blocked by default.
- Export Options — Save query results to CSV, Excel, or SQLite with a single flag.
- Dry-Run & Explain Modes — Preview generated SQL and execution plan without affecting the database.
Use cases
- Generate a sales report for the last quarter without writing SQL
- Find inventory items low in stock using conversational prompts
- Export query results to CSV for downstream analysis
- Validate complex joins by previewing generated SQL first
TESH-Query FAQ
How do I install TESH-Query?+
Run `pip install teshq` or include extras like `teshq[postgres]` for specific drivers.
Which databases are supported?+
PostgreSQL, MySQL, SQLite, MSSQL, MariaDB, Oracle, and any DB supported by SQLAlchemy.
Can I see the generated SQL before it runs?+
Yes, use the `--dry-run` or `--explain` flags to preview and explain the query.
How are credentials stored?+
All configuration, including DB and LLM credentials, are saved securely in `~/.teshq/`.
Is the tool free to use?+
TESH-Query is open-source under the MIT License and can be installed via pip at no cost.
Can I integrate TESH-Query into my own Python app?+
Yes, the Python SDK provides a `TeshQuery` class for programmatic NL-to-SQL queries.
Summarized by DevHunt from pypi.org · Oct 8, 2026. Details may change; check the official site.













