Ask in Slack
Mention the bot with a question. Cassis replies in the thread with the answer, SQL, and a link to the full chat.
An admin must first install Slack and map the channel to a project. Then mention the bot with the question.
@Cassis who won the most races in 2023?
@you Max Verstappen won 19 of 22 races in 2023, far ahead of Pérez (2) and Sainz (1).
sql: SELECT d.full_name, COUNT(*) FROM public.results r
JOIN public.races ra ON r.race_id = ra.race_id
WHERE ra.year = 2023 AND r.position = 1 GROUP BY 1
results: 3 rows
JOIN public.races ra ON r.race_id = ra.race_id
WHERE ra.year = 2023 AND r.position = 1 GROUP BY 1
results: 3 rows
Open in Cassis ↗ 👍 / 👎
@Cassis now break it down by constructor.
@you Red Bull Racing 860, Mercedes 409, Ferrari 406, McLaren 302, and six more.
results: 10 rows attached as results.csv
Continue in the thread
- Follow-ups
- Reply by mentioning the bot again. The thread keeps the Cassis conversation context.
- Plans
- When an assumption needs your decision, use Run plan to approve it or Change assumptions to correct it.
- Ratings
- Rate the answer in the thread. The feedback reaches the same review queue as web-app feedback.
- Large results
- Cassis attaches
results.csvand keeps a preview inline.
Only the person who started the thread can continue it, approve its plan, or rate its answers. Other people start their own thread.
Slack is for asking, not editing the ontology. Answers come from the same published version used by the web app and MCP clients.