Help center / Guides / Troubleshooting checklist
Guide • July 28, 2026

Before you change anything, describe what is true.

This checklist helps operators preserve evidence and choose a diagnostic action that proves or eliminates one meaningful layer.

1. Write the symptom

Expected

What exact task and result should occur?

Observed

What appears, fails, changes, or remains unavailable?

Impact

Who is blocked, what work is affected, and is there a safe workaround?

2. Preserve the baseline

Record timestamps, versions, platform, workspace, data source, entry type, error wording, and recent changes before clearing caches, resetting profiles, reinstalling, or mass-editing.

  • Redact secrets and personal data
  • Capture configuration, not credentials
  • Note last known successful use
  • Save logs through approved channels
  • Do not test against critical targets without authorization

3. Choose one comparison

ComparisonQuestion
Another userIs the issue role, profile, or local-client specific?
Another entryIs configuration, credential, or target specific?
Another platformIs behavior client implementation dependent?
Direct approved testIs the target reachable outside the saved path?

4. Record what the result proves

Avoid “it worked” as the entire conclusion. State the variable changed, observed result, what layer becomes more or less likely, and the next safe step.

Review the triage method

Need a second set of eyes?

Send a redacted evidence summary, never secrets.

Prepare an inquiry