Or you can use an LLM that uses the github MCP to get all February 2026 SUMO Thunderbird Desktop Questions from my CSV files and then you won't need to learn SQLite. I think this should be YES...AND... One needs both at least in 2026 :-) ! Sample prompt follows:
- Draft created: Mar 15, 2026 21:48 (UTC).
Using the available tools, graph tags in January 2026 (filenames: 2026-01*thunderbird-creator-answers-desktop-all-locales.csv) versus February 2026 (filenames: 2026-02*thunderbird-creator-answers-desktop-all-locales.csv) located in the 2026 directory of the thunderbird/github-action-thunderbird-aaq repo. The tags are semi-colon delimited. Please ignore the 'thunderbird' tag.- (i bet you can do this with a local model so you don’t have to destroy the environment more with a cloud model)
- It’s 2026 you still need to learn SQL to do basic queries if you work with data Maybe by 2030 local models will be so good you won’t have to?
- Previously: March 10, 2026: How to get all February 2026 SUMO Thunderbird Desktop Questions from my CSV files (updated via a github action once per hour)l use ‘select all’ to unify the 28 CSVs once per day and TIL you can add fields using ‘select’