Issue

The issue pipeline connects user-reported bugs to automated fixes. Users click "Report an Issue" in the UI, a GitHub issue is created, and the report agent processes it.

Report an Issue (User-Facing)

Every hogwarts page has a "Report an Issue" link or icon. When clicked, a dialog opens with a textarea. On submit, a GitHub issue is created in databayt/hogwarts with the report label.

Where It Appears

Component

src/components/report-issue.tsx

Two variants:

• text (default) — underlined "Report an issue" link that opens the dialog
• icon — Bug icon (lucide-react) that opens the same dialog

The dialog contains only a textarea and submit button. Bilingual via useDictionary().

Server Action

src/lib/actions/report-issue.ts

On submit:

1. Reads GITHUB_PERSONAL_ACCESS_TOKEN from env
2. Reads auth session via auth() (for reporter identification)
3. Creates GitHub issue with report label
4. Posts acknowledgment comment: "Received. This report is queued for automated review and fix."

Issue Body Format

Copy<user's description>
 
---
 
**Page**: https://kingfahad.databayt.org/en/admissions
**Time**: 2026-04-02T08:42:26Z

Title is truncated to 80 characters from the description.

i18n

Translation keys in en.json / ar.json under reportIssue:

Report Agent (Auto-Fix)

The report agent processes open report-labeled issues. Triggered manually by typing report in Claude Code, or by the session-start hook.

Trigger

Copy# Manual
report
report hogwarts
 
# Session-start hook (automatic)
# Checks hogwarts + kun for open report issues on every session start

Pipeline

READ → LOCATE → CONTEXT → VALIDATE → FIX → BUILD → PUSH → CLOSE

Validation Gate

Before writing any code, the agent answers three questions:

1. Is it a real bug? — Can it reproduce from description + URL? If not → cannot-reproduce label
2. Is it aligned? — Does it contradict ISSUE.md planned work? If feature request → needs-human label
3. Is the fix safe? — Could it break shared code? When in doubt → needs-human label

Escalation

Agent Definition

.claude/agents/report.md

Close Comment Format

CopyFixed in abc1234.
 
**What was wrong**: <root cause>
**What was fixed**: <change summary>
**Files changed**: <list>
**Verified**: build passes, auto-deploying to production

Scheduled Trigger (Paused)

A remote trigger exists that runs the report agent every 2 hours in Anthropic's cloud. Currently paused — using manual report keyword until the feature stabilizes.

To re-enable when ready:

Copy# In Claude Code
schedule update trig_01MFLRtUTfMMNTDGBBGQtZLq --enabled true

Session-Start Hook

Every time Claude Code opens, the session-start hook checks for open report issues:

Copy# In ~/.claude/settings.json (SessionStart hook)
gh issue list --label report --state open --repo databayt/hogwarts
gh issue list --label report --state open --repo databayt/kun

If found, the agent processes them immediately per the rule in session-start.md.

Full Pipeline Flow

User clicks "Report an issue"
    ↓
Dialog opens → user types description → Submit
    ↓
Server action creates GitHub issue (label: report)
    ↓
Acknowledgment comment posted: "Received. Queued for review."
    ↓
Next Claude Code session → hook detects → agent processes
    OR
User types "report" → agent processes manually
    OR
Scheduled trigger (when enabled) → agent processes every 2h
    ↓
Agent: READ → LOCATE → CONTEXT → VALIDATE → FIX → BUILD → PUSH
    ↓
Issue closed with structured comment
    ↓
Vercel auto-deploys fix to production

Files

Hogwarts

Kun (Engine)

Issue Creation (CLI)

Issues can also be created from Claude Code or Slack, not just the dialog.

Usage

Copy# From Claude Code
/issue "Login crashes on Safari mobile" --repo hogwarts
 
# From keyword
issue "Add dark mode to dashboard"

Repo Mapping

Labels

Priority

Type

Issue Body Template (CLI)

Copy## Context
Why this matters — user impact, business context, or technical debt.
 
## Action
- [ ] Step 1
- [ ] Step 2
- [ ] Step 3
 
## Verification
How to confirm it's done — test commands, URLs to check, expected behavior.