Skip to content

Latest commit

 

History

History
60 lines (38 loc) · 1.79 KB

File metadata and controls

60 lines (38 loc) · 1.79 KB

Exercise 02 — Analyze Logs with #file

ShopSphere Workshop · Time: 5 min · Feature: Copilot Chat + #file

MANDATORY — First Copilot skill: attach a file to Chat and extract meaning from logs.


Objective

Use #file to point Copilot directly at the production log file and get a structured error analysis — without reading the file yourself.


How #file Works

In Copilot Chat, type #file: followed by a path (or pick from the dropdown).
Copilot reads the full file content inline and reasons over it.

#file:observability/production-logs.txt

This is faster than copy-pasting — and works for any file in the workspace.


Step 1 — Open Copilot Chat

Press Ctrl+Alt+I (Windows/Linux) or Cmd+I (macOS)


Step 2 — Paste This Prompt

📋 COPY AND PASTE INTO COPILOT CHAT:

Analyze the log file #file:observability/production-logs.txt

For each distinct error pattern found:
1. Classify the error type (e.g. NullPointerException, swallowed exception, timeout)
2. Identify which user IDs are affected and why
3. Estimate the blast radius — what % of users does this impact?
4. Assign a severity: P0 / P1 / P2

Which issue should I investigate first, and why?

What to Look For in Copilot's Response

  • Does it identify both error patterns (null crash AND silent payment failure)?
  • Does it distinguish user IDs that always fail vs. those that sometimes fail?
  • Does it note what's absent from the logs (missing stack trace, no correlation ID)?

Checkpoint ✓

Copilot identifies at least two distinct error patterns and correctly rates the null crash as higher priority (50% blast radius vs. intermittent payment failures).

Next: Exercise 03 — Map Architecture with Chat