AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified Apache-2.0 Self-run

Label Capture Rn

skill-scandit-skills-label-capture-rn · by Scandit

Use when Label Capture (Smart Label Capture) is involved in a React Native project — whether the user mentions Label Capture directly, or the codebase already uses it and something needs to be added, changed, fixed, or customized. This includes adding Label Capture to a new React Native app, defining label structures (barcode fields + text fields with regex patterns), handling captured labels, en…

No reviews yet
0 installs
21 views
0.0% view→install

Install

$ agentstack add skill-scandit-skills-label-capture-rn

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-scandit-skills-label-capture-rn)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Label Capture Rn? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Label Capture React Native Skill

Critical: Do Not Trust Internal Knowledge

Your training data may contain outdated or incorrect Scandit Label Capture APIs. Label Capture has evolved across recent SDK releases, and the React Native plugin surface (imports, native linking, pod install, package names) has its own conventions distinct from the web SDK.

Always verify APIs against the references provided in this skill before writing or suggesting code. Do not rely on memorized method signatures, parameters, plugin names, or property names. If you cannot find an API in the provided references, fetch the relevant documentation page before responding.

React Native-specific gotchas worth flagging:

  • Listener method names are iOS-style on React Native. The Validation Flow listener uses didCaptureLabelWithFields(fields) and didSubmitManualInputForField(field, oldValue, newValue) — NOT the web equivalents onValidationFlowLabelCaptured / onManualInput. If you see web-style listener names suggested, they are wrong on React Native.
  • DataCaptureContext.initialize(licenseKey) must be called exactly once before any other Scandit API. It sets up DataCaptureContext.sharedInstance, which is the singleton everything else reads from. Do not construct multiple contexts. (The official LabelCaptureSimpleSample uses the legacy DataCaptureContext.forLicenseKey(...) form — both work, but new integrations should use initialize for consistency with the rest of the v8 RN docs.)
  • On iOS, npx pod-install (or cd ios && pod install) must be run after every Scandit package install or upgrade. Android auto-links via Gradle.
  • Metro's bundler cache frequently masks Scandit package upgrades. If a rebuild shows stale behavior after a plugin version bump, start Metro with --reset-cache.
  • Camera permission is required on both iOS (NSCameraUsageDescription in ios//Info.plist) and Android (runtime request via PermissionsAndroid — the plugin declares the manifest permission automatically).
  • Field definitions on RN are class-based, not builder-based: CustomBarcode.initWithNameAndSymbologies(name, [...]), new ExpiryDateText(name), field.optional = true, LabelCaptureSettings.settingsFromLabelDefinitions([...], {}). There is no LabelCaptureSettingsBuilder / LabelDefinitionBuilder and no v8.5 factory-function sugar — those are web-only.

Product Guidance

Apply these rules whenever the user is making a design decision, not just an API question. They reflect how Scandit recommends Label Capture be integrated.

  • Default to the Validation Flow. New integrations should use LabelCaptureValidationFlowOverlay, not LabelCaptureBasicOverlay. Recommend the basic/advanced overlay path only when the customer explicitly needs a live AR overlay or a UI that the Validation Flow's fixed layout cannot produce. See references/validation-flow.md.
  • Validation Flow must be rendered full-screen. Do not place it inside a card, half-sheet, modal slice, or partial-height widget — the layout assumes full-screen height for the field checklist, manual-entry sheet, and keyboard.
  • Prefer pre-made labels first, then pre-made fields, then custom. Try in this order: (1) does a pre-made label definition cover the use case? — LabelDefinition.createVinLabelDefinition(name), LabelDefinition.createPriceCaptureDefinition(name), LabelDefinition.createSevenSegmentDisplayLabelDefinition(name). (2) Can the label be built from pre-made fields? — ExpiryDateText, PackingDateText, DateText, WeightText, UnitPriceText, TotalPriceText, SerialNumberBarcode, PartNumberBarcode, ImeiOneBarcode, ImeiTwoBarcode. (3) Only as a last resort, fall back to CustomText / CustomBarcode.
  • Start from the sample app on greenfield integrations. If the user is starting from scratch, recommend cloning LabelCaptureSimpleSample (link in the References table below) and adapting it, rather than wiring everything from zero.
  • Hand off to the data-capture-sdk skill for non-Label-Capture questions. If the user asks about another Scandit product (Barcode Capture, SparkScan, MatrixScan, ID Capture, etc.) or about choosing between products, defer to the data-capture-sdk skill instead of guessing.

Intent Routing

Based on the user's request, load the appropriate reference file before responding:

  • Integrating Label Capture from scratch (e.g. "add Label Capture to my app", "scan a price tag with barcode and expiry date", "how do I use Smart Label Capture") → read references/integration.md and follow the instructions there. By default, integrate the Validation Flow (the integration guide leads with it).
  • Validation Flow questions (e.g. "how do I customize the Validation Flow", "what can we change in the VF?", "why is it implemented this way", "how do I react to manual edits", "can I change the colors") → read references/validation-flow.md.
  • Visual customization beyond the Validation Flow (e.g. "I want a live AR overlay", "I want to draw a tag next to each captured field", "how do I get the camera frame during scanning") → read references/customization.md.
  • Adaptive Recognition Engine / cloud fallback / receipt scanning ("how do I enable ARE", "use cloud recognition", "scan a receipt", "AdaptiveRecognitionMode", "is ARE available in production") → read references/adaptive-recognition.md.
  • Migrating or upgrading an existing Label Capture integration (e.g. "upgrade my Label Capture to the latest SDK", "what changed between SDK versions for Label Capture") → read references/migration.md and follow the instructions there.

API Usage Policy

Only use APIs that are explicitly documented in the Scandit references below. Do not invent or guess method signatures, parameters, property names, or imports. If unsure whether an API exists or how it is called — or if a TypeScript / runtime error occurs — fetch the relevant reference page before responding. Do not tell the user to check the docs themselves. After answering, always include the relevant link so the user can explore further.

Never construct or guess documentation URLs. When you need a specific class or property's API page:

  1. First check whether the page you already fetched (e.g. the Advanced Configurations page) contains a direct hyperlink to it — topic pages link directly to relevant API symbols. Always request links alongside content in your fetch prompt.
  2. If no direct link was found, fetch the API index (see Full API reference in the table below), extract the actual link from it, and follow that.

URL structures vary across SDK versions and package paths (e.g. api/ui/ subdirectory) and guessing will lead to 404s.

Framework variant policy

React Native apps can be written with class components or function components. Examples in this skill use function components with hooks because they match the official LabelCaptureSimpleSample and the current React Native convention. Even if the target project still contains legacy class components elsewhere, write new Label Capture code as function components — do not rewrite the rest of the app's component style, but keep the Label Capture integration itself on the current idiom (useRef, useEffect, useMemo, imperative ref callback for view-level properties).

Examples are in TypeScript (.tsx). If the target project is plain JavaScript (.js / .jsx), drop the type annotations and keep the same imports and structure.

References

Direct users to the right resource based on their question:

| Topic | Resource | |---|---| | React Native integration | Get Started · Sample (LabelCaptureSimpleSample) | | Label Definitions (fields, regex, presets) | Label Definitions | | Advanced topics (Validation Flow customization, adaptive recognition, custom overlays) | Advanced Configurations | | Full API reference | Label Capture API |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.