# Track and Triage Issues Automatically

> AI-powered issue tracking with automatic categorization and intelligent routing. Capture issues from email, Slack, webhooks, or API calls with smart deduplication.

Canonical: https://issues.do/

## In code

```ts
import { Issues } from 'issues.do'

// Auto-triage and route issues with AI
on.Issue.created($ => {
  const analysis = $.AI.analyze.Issue({
    text: $.title,
    categories: ['bug', 'feature', 'support', 'security']
  })
  $.assign.to.Team({
    team: analysis.category === 'security' ? 'Security' : 'Engineering',
    priority: analysis.severity
  })
  $.Slack.send.Message({ channel: `#${analysis.category}`, text: $.title })
})
```

## Frequently asked questions

### How does AI-powered triage work?

Issues.do uses LLMs (GPT, Claude, Gemini) to analyze issue titles and descriptions, extract key information, determine severity (critical/high/medium/low), categorize by type (bug/feature/support/security), suggest relevant labels, and route to appropriate teams. AI learns from your past issues to improve accuracy over time. Reduce manual triage time by 80%—most issues are categorized correctly without human review.

### Can issues.do deduplicate similar issues automatically?

Yes. Semantic similarity detection identifies duplicate or related issues using embeddings and vector search. When a new issue matches existing ones (>85% similarity), it's automatically linked or merged. Configurable thresholds let you control sensitivity. Users are notified of related issues when submitting new ones. Deduplication reduces issue volume by 30-40% in typical support workflows.

### What sources can issues.do capture from?

Email (support@yourcompany.com), Slack channels, web forms, API endpoints, GitHub webhooks, Sentry error reports, customer chat, mobile app crashes, and custom integrations. All sources feed into a unified issue tracker. Each source can have custom parsing rules and routing logic. Issues include full context (user info, environment, logs) automatically extracted from the source.

### How do SLA tracking and escalation work?

Define SLAs per issue priority (e.g., critical issues require response in 1 hour, resolution in 4 hours). Automatic escalation when SLAs are at risk—notify managers, reassign to senior engineers, or page on-call team. Track SLA compliance per team, customer, and time period. Dashboards show breach rates, average response times, and team performance. Pause SLA timers outside business hours or during maintenance windows.

### Can customers track their issues in a portal?

Yes. White-label customer portal shows issue status, updates, and resolution timeline. Customers can submit new issues, attach files, comment on existing issues, and subscribe to notifications. SSO integration for enterprise customers. Embed portal in your app via iframe or use our hosted solution. Full audit trail shows customer when issues were acknowledged, assigned, and resolved. GDPR-compliant data exports available.

## Part of Platform.do

issues.do is one of the primitives of Platform.do, in the Business layer (goals, plans and the work itself). Every .do shares one grammar, Business-as-Code, and one SDK.

- What is Business-as-Code? https://platform.do/business-as-code
- Platform.do: https://platform.do/
- Docs: https://docs.platform.do