AI Support

Support

Purpose

Structured customer support investigation. Organizes findings by ticket, links to relevant code and PRs, and builds a searchable knowledge base of resolved issues.

Trigger

• Command: /ai-support start <ticket-id> or /ai-support find [query]
• Context: customer-reported issue, support ticket investigation, escalation from support team.

When to Use

• Investigating a customer-reported bug or behavior
• Reproducing an issue from a support ticket
• Documenting resolution for future reference
• Escalation requiring code-level investigation

When NOT to Use

• Production incidents -- use /ai-postmortem
• Internal bugs found during development -- use /ai-debug
• Feature requests -- use /ai-triage

Modes

start -- New investigation

1. Create structure -- create .ai-engineering/support/{date}/{ticket-id}/ directory.
2. Scaffold investigation -- create investigation.md from template:

# {ticket-id}: {title}

**Date**: YYYY-MM-DD
**Customer**: {name/org if known}
**Status**: investigating | resolved | escalated
**Priority**: p1 | p2 | p3

## Issue
{Customer's description -- verbatim or summarized}

## Environment
- Product version:
- OS/Platform:
- Configuration:

## Steps to Reproduce
1. {Step}
2. {Step}
3. {Expected vs actual behavior}

## Findings
{Investigation results, root cause analysis}

## Resolution
{Fix applied, workaround provided, or escalation path}

## Related
- Code: {file paths}
- PR: {links}
- Notes: {links to /ai-note entries}

3. Investigate -- explore codebase for relevant code paths, check recent changes to affected areas, review error patterns.
4. Update -- keep investigation.md current as findings emerge.

find [query] -- Search investigations

1. Search -- scan .ai-engineering/support/ directories for matching content.
2. Rank -- prioritize by recency, then relevance.
3. Present -- list ticket-id, date, title, status, and resolution summary.

Procedure for Investigation

1. Reproduce -- attempt to reproduce the issue locally using the reported steps.
2. Isolate -- narrow down to the specific code path, configuration, or data condition.
3. Root cause -- identify why the behavior occurs (bug, misconfiguration, edge case, expected behavior).
4. Resolve -- one of:
   • Fix: create a PR via /ai-pr and link it in the investigation
   • Workaround: document the workaround steps
   • Escalate: mark as escalated with reason and target team
   • Won't fix: document rationale

Quick Reference

/ai-support start TICKET-4521         # start investigation
/ai-support start SUP-123             # any ticket ID format works
/ai-support find timeout              # search past investigations
/ai-support find                      # list all investigations

Storage

• Location: .ai-engineering/support/{YYYY-MM-DD}/{ticket-id}/investigation.md
• Organized by date for natural chronological browsing

$ARGUMENTS