The task
Teams need the same numbers every day or every week: reports for key clients, summaries and alerts, plan and operations reports.
Before, someone pulled the data and built each report by hand.
Client teams, operations and planning
Reports that build themselves on a schedule
Agents build daily and weekly reports on their own and post them where the team works.
Key clients: daily reportEvery day at 08:00
Every morning, post yesterday’s numbers for our key clients to #client-reports. Add a line on anything that moved more than 10%.
Data warehouse Run queries · 6
Code Build the table
Team chat Post to #client-reports
Posted to #client-reports. 2 clients moved more than 10% day over day: Northwind is up 14% after a promotion, Lumen Retail is down 12% with no orders after 18:00.
ClientOrdersRevenuevs day before
Northwind1,284$96,300+14%
Lumen Retail842$51,900−12%
Harbor Foods1,015$74,200+2%
Message…
AgentClient Reports
ModelGPT 6.1 Sol
ReasoningAuto
Screens show sample data. The case is real use in the group that built Sigma.
How the agent does it
- A person describes the report once: what to count, how to lay it out, where to post it and when.
- On schedule, the agent pulls data from the warehouse and other systems, does the counting and writes a short text with tables.
- It posts the report to the team’s channel and keeps it in the run history. People get a notification.
- Every run is a chat. People can open it and ask a follow-up question.
- A report can run as a service account, not a person. It keeps going when someone is away or leaves.
What changed
- Nobody builds these reports by hand.
- They run on their own. A person steps in only to change something or ask a question.
In numbers
Frequency
5,055scheduled runs in 30 days
Last 30 days
Finished
99.4%of scheduled runs went to the end without a person stepping in
An earlier 30-day period
Production data inside the group, read-only. A run is finished when it ends with the status “completed”.
More from our own use
Finance, sales, product and operations
Answers from company data, without waiting for an analyst
People ask about the company’s data in plain words. The agent writes the queries and answers with a table.
Read the case Technical supportA first look at every incident ticket
The agent picks up each new incident ticket, gathers the facts and posts the likely cause to the support chat.
Read the case Customer supportFirst line of customer support
In one of the group’s products, the agent takes new customer requests, looks up the account and brings in the teams it needs.
Read the case