Settings Reference

Settings Reference

All configuration lives under the unified Settings page in the Lastest UI.

---

GitHub

Connect your GitHub account for:

• OAuth login
• Repository sync
• PR comments with build results
• Webhook-triggered builds

See [GitHub Integration] for details.

---

GitLab

Connect your GitLab account for:

• OAuth login (self-hosted instances supported)
• MR comments with build results
• Webhook-triggered builds

See [GitLab Integration] for details.

---

Google Sheets

Connect to Google Drive to use spreadsheet data as test data sources.

See [Google Sheets Integration] for details.

---

Playwright

Setting	Description
Browser type	Chromium, Firefox, or WebKit
Viewport	Width and height of the browser window
Headless mode	Run without visible browser (includes shell mode)
Selector priority	Order for selector strategies (data-testid, id, role, etc.)
Recording engine	Custom recorder or Playwright Inspector
Animation freezing	Disable CSS animations during capture
Screenshot delay	Wait time before capturing
Max parallel tests	Maximum concurrent test execution

---

Stabilization

Setting	Description
Network idle	Wait for network activity to settle
DOM stability	Wait for DOM mutations to stop
Timestamp freezing	Replace Date.now() and new Date() with fixed values
Random seeding	Seed Math.random() for consistent outputs
Third-party blocking	Block third-party domains with configurable allowlist
Font loading	Wait for webfonts to load, or disable them entirely
Loading indicator hiding	Auto-hide spinners and loading states
Cross-OS consistency	Bundled fonts + Chromium flags for identical screenshots
Burst capture	Multi-frame instability detection
Auto-mask dynamic content	Automatically mask timestamps, UUIDs, and relative times

See [Stabilization Features] for details.

---

Environment

Setting	Description
Server startup	Manual vs auto-start
Health check URLs	URLs to verify your app is running before testing

---

Diff Sensitivity

Setting	Description
Diff engine	pixelmatch, SSIM, or Butteraugli
Text-region-aware	OCR-based two-pass comparison
Pixel threshold	Number of pixels to trigger change
Percentage threshold	Percentage of pixels to trigger change
Page shift detection	Detect vertical content shifts
DOM diff fallback	Run a structural DOM diff alongside pixel comparison when results are inconclusive
Per-step ignore regions	Ignore regions are scoped to the screenshot they were drawn on (not the whole test)

See [Visual Diffing] for details.

---

AI

Setting	Description
Test generation provider	Which AI provider to use for generating/fixing tests
Diff analysis provider	Separate provider for diff classification
API keys	Provider-specific keys
Model	Specific model to use
Custom instructions	Extra context for AI prompts
OpenAI support	OpenAI API key and model selection
Ollama support	Local model configuration

See [AI Configuration] for details.

---

Notifications

Channel	Configuration
Slack	Webhook URL
Discord	Webhook URL
Custom webhook	URL, HTTP method, custom headers
GitHub PR	Automatic (via GitHub integration)
GitLab MR	Automatic (via GitLab integration)

See [Custom Webhooks] for webhook payload details.

---

Branches

Setting	Description
Baseline branch	Which branch holds the source-of-truth baselines
Scanning branch	Which branch to scan for routes

---

AI Logs

Audit trail of all AI requests -- last 50 entries. View what was sent and received.

---

Testing Templates

One-click preset configurations for common app types:

Template	Optimized For
SaaS/Dashboard	Data-heavy UIs, tables, charts
Marketing Website	Static pages, animations, responsive
Canvas/WebGL	Non-DOM rendering, visual-only comparison
E-commerce	Product pages, carts, checkout flows
Documentation	Text-heavy, code blocks, navigation
Mobile-First	Small viewports, touch interactions
SPA	Client-side routing, dynamic content
CMS	Content editing, WYSIWYG, media

---

Setup & Teardown

Setup

Default repository-wide multi-step setup sequences:

• Playwright (browser) scripts
• API (HTTP) scripts
• Test-as-setup scripts

Teardown

Default repository-wide teardown sequences:

• Per-test overrides available (skip or add steps)
• Teardown errors are non-blocking

---

Schedules

Setting	Description
Preset schedules	Daily, weekly, hourly, every 6 hours, every 15 minutes
Custom cron	Any valid cron expression
Branch targeting	Run against a specific branch
Enable/disable	Toggle schedules on and off

See [Scheduled Runs] for details.

---

Vars

The repository-wide test data tab. Three kinds of variables coexist:

Source	Example
Static	Hand-typed key/value pairs
AI-generated	Preset-driven (name, email, IBAN, address...) -- AI fills realistic values per run
Google Sheets	A pinned column from a connected sheet

See [AI Configuration] (AI vars) and [Google Sheets Integration] (sheet-backed vars).

---

API Tokens

Long-lived Bearer tokens for programmatic + MCP access. See [API Tokens].

---

Account

Per-user controls:

• Email preferences -- which build / review notifications arrive in your inbox
• Email unsubscribe -- one-click opt-out of all transactional and digest mail
• Self-serve account deletion -- GDPR-compliant delete that removes user data, anonymises authored test history, and revokes outstanding tokens
• GDPR opt-in -- recorded at signup; surfaced here for audit and re-consent

---

Users (Admin Only)

• View team members
• Send invitations
• Manage roles (owner, admin, member, viewer)

---

Runners (Admin Only)

• Register new remote runners
• View connected runners
• Monitor runner status and system info
• Configure parallel limits