lizhiyao/sentry-miniapp/.agents/skills/sentry-miniapp-sdk/SKILL.md
sentry-miniapp-sdk
Full Sentry SDK setup for Mini Programs — error monitoring, tracing, offline cache, source maps. Supports WeChat, Alipay, ByteDance, Baidu, QQ, DingTalk, Kuaishou and cross-platform frameworks (Taro / uni-app).
- Source repository stars
- 677
- Declared platforms
- 0
- Static risk flags
- 3
- Last source update
- 2026-08-28
- Source checked
- 2026-08-28
Decision brief
What it does: where it fits
Set up Sentry error monitoring, performance tracing, and offline caching in mini program projects using sentry-miniapp — a community SDK built on @sentry/core v10.
Not for
- Tasks that require unconfirmed production actions or broad system permissions.
- Environments where the pinned source and install steps cannot be inspected.
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/lizhiyao/sentry-miniapp --skill ".agents/skills/sentry-miniapp-sdk"Inspect the Agent Skill "sentry-miniapp-sdk" from https://github.com/lizhiyao/sentry-miniapp/blob/cf13bb46035082d78988692af3251430e9c0ccc0/.agents/skills/sentry-miniapp-sdk/SKILL.md at commit cf13bb46035082d78988692af3251430e9c0ccc0. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
Phase 1: Detect
Run these commands to understand the project:
Run these commands to understand the project: - 02
Phase 2: Recommend
Present this recommendation based on detection results:
✅ Error Monitoring — always. Automatic capture of onError, onUnhandledRejection, onPageNotFound, onMemoryWarning✅ Performance Tracing — always. Navigation timing, render performance, resource loading, custom spans⚡ Source Map Upload — when deploying to production. Maps minified stack traces back to source code - 03
Phase 3: Guide
Review the “Phase 3: Guide” section in the pinned source before continuing.
Review and apply the “Phase 3: Guide” source section. - 04
Step 1: Install
Review the “Step 1: Install” section in the pinned source before continuing.
Review and apply the “Step 1: Install” source section. - 05
Step 2: Initialize
Create or modify the app entry point. Sentry.init() must be called before App().
Create or modify the app entry point. Sentry.init() must be called before App().
Permission review
Static risk signals and limitations
Runs scripts
The documentation asks the agent to run terminal commands or scripts.
# npmRuns scripts
The documentation asks the agent to run terminal commands or scripts.
npm install sentry-miniappNetwork access
The documentation includes network, browsing, or remote request actions.
dsn: 'https://<key>@<org>.ingest.sentry.io/<project>',Network access
The documentation includes network, browsing, or remote request actions.
dsn: 'https://<key>@<org>.ingest.sentry.io/<project>',Reads files
The documentation asks the agent to read local files, directories, or repositories.
Walk through features one at a time. Load the corresponding reference file:Evidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 92/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 677 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- lizhiyao/sentry-miniapp
- Skill path
- .agents/skills/sentry-miniapp-sdk/SKILL.md
- Commit
- cf13bb46035082d78988692af3251430e9c0ccc0
- License
- MIT
- Collected
- 2026-08-28
- Default branch
- master
View the original SKILL.md
Sentry Mini Program SDK Setup
Set up Sentry error monitoring, performance tracing, and offline caching in mini program projects using sentry-miniapp — a community SDK built on @sentry/core v10.
Invoke This Skill When
- User mentions mini program, miniapp, 小程序, WeChat, Alipay, ByteDance, Taro, uni-app alongside Sentry
- User wants to add error monitoring or performance tracking to a mini program
- User asks about Sentry support for WeChat/Alipay/ByteDance mini programs
- User imports or references
sentry-miniappin their project
Note: SDK versions and APIs reflect the current sentry-miniapp docs. Always verify against the sentry-miniapp README before implementing.
Phase 1: Detect
Run these commands to understand the project:
# Detect mini program platform
ls app.json project.config.json mini.project.json project.tt.json project.swan.json project.qq.json 2>/dev/null
cat app.json 2>/dev/null | head -20
# Detect framework (Taro / uni-app / native)
cat package.json 2>/dev/null | grep -E '"@tarojs/|"@dcloudio/uni-|"sentry-miniapp"'
# Check for existing Sentry SDK
grep -r "sentry" package.json 2>/dev/null
grep -r "Sentry.init" app.js app.ts src/app.js src/app.ts 2>/dev/null
# Detect entry point
ls app.js app.ts src/app.js src/app.ts 2>/dev/null
# Detect package manager
ls yarn.lock pnpm-lock.yaml package-lock.json 2>/dev/null
What to determine:
| Question | Impact |
|---|---|
| Which mini program platform? | Determines platform option and API differences |
| Native or cross-platform framework (Taro/uni-app)? | Determines init pattern and build config |
Is sentry-miniapp already installed? | Skip install if present, check version |
| Where is the app entry point? | Determines where to place Sentry.init() |
| Is there a build tool config (webpack/vite)? | Needed for Source Map setup |
Platform Detection Guide
| File Present | Platform |
|---|---|
project.config.json | WeChat (微信) |
mini.project.json | Alipay (支付宝) |
project.tt.json | ByteDance (字节跳动) |
project.swan.json | Baidu (百度) |
project.qq.json | |
@tarojs/* in package.json | Taro (cross-platform) |
@dcloudio/uni-* in package.json | uni-app (cross-platform) |
DingTalk and Kuaishou do not have a stable project-file signal in this guide. Confirm them from platform globals, framework configuration, dependencies, or the user's stated target instead of guessing from a filename.
Phase 2: Recommend
Present this recommendation based on detection results:
Recommended (core coverage):
- ✅ Error Monitoring — always. Automatic capture of
onError,onUnhandledRejection,onPageNotFound,onMemoryWarning - ✅ Performance Tracing — always. Navigation timing, render performance, resource loading, custom spans
Recommended for production:
- ⚡ Source Map Upload — when deploying to production. Maps minified stack traces back to source code
- ⚡ Offline Cache — when users are on unreliable networks. Caches events locally, retries when connectivity returns
Optional:
- ⚡ Distributed Tracing — when mini program calls backend APIs. Injects
sentry-trace/baggageheaders, and can optionally add W3Ctraceparent, to link frontend and backend spans - ⚡ User Feedback — when you want to collect user-reported issues via
Sentry.captureFeedback()
| Feature | Recommend when... |
|---|---|
| Error Monitoring | Always — zero-config automatic exception capture |
| Performance Tracing | Always — automatic page/network/render performance |
| Source Map | Production — required to read minified stack traces |
| Offline Cache | Weak networks — mobile users, rural areas |
| Distributed Tracing | API calls — mini program talks to backend services |
| User Feedback | User-facing — collect bug reports from end users |
Phase 3: Guide
Step 1: Install
# npm
npm install sentry-miniapp
# yarn
yarn add sentry-miniapp
Step 2: Initialize
Create or modify the app entry point. Sentry.init() must be called before App().
Native Mini Program (app.js)
const Sentry = require('sentry-miniapp');
Sentry.init({
dsn: 'https://<key>@<org>.ingest.sentry.io/<project>',
release: '[email protected]',
environment: 'production',
});
App({
// Your app config...
// No need to manually call Sentry in onError — SDK handles it automatically
});
Taro (app.js or app.ts)
import * as Sentry from 'sentry-miniapp';
Sentry.init({
dsn: 'https://<key>@<org>.ingest.sentry.io/<project>',
release: '[email protected]',
environment: 'production',
});
// Taro App component follows...
uni-app (App.vue or main.js)
const Sentry = require('sentry-miniapp');
Sentry.init({
dsn: 'https://<key>@<org>.ingest.sentry.io/<project>',
release: '[email protected]',
environment: 'production',
});
Step 3: Configure Platform (if needed)
The SDK auto-detects the platform at runtime. When multiple platform globals are present, it uses platform-specific host names, data paths, and App IDs to disambiguate them. Set platform only when those runtime signals are unavailable, conflicting, or the detected event label still does not match the release target.
If you need to override the event's miniapp platform label:
Sentry.init({
dsn: '...',
platform: 'bytedance', // 'wechat' | 'alipay' | 'bytedance' | 'qq' | 'swan' | 'dingtalk' | 'kuaishou'
// Note: Baidu's global object is `swan`, so use 'swan' (there is no 'baidu' value).
});
The configured value is written to contexts.miniapp.platform; it does not switch the underlying runtime API. The top-level Sentry event platform always remains the standard value javascript so stack parsing, grouping, and Source Maps follow official JavaScript SDK semantics. Error handlers, requests, and storage continue using the compatible global detected automatically.
Step 4: Add User Context
// After user login
Sentry.setUser({
id: 'user-123',
username: 'zhang_san',
});
// Add custom tags
Sentry.setTag('page', 'payment');
Sentry.setContext('order', { orderId: '2024001', amount: 99.9 });
Minigame (小游戏)
WeChat / ByteDance minigames have no App() / Page() / routing, so the page-based integrations (PageBreadcrumbs, Session) cannot work there. The SDK detects minigames via crossPlatform.isMinigame() and switches to dedicated integrations — enabled by default only in a minigame runtime, and a safe no-op in a regular mini program:
MinigameIntegration(enableMinigameLifecycle) — reads the launch scene (scene/path/query) fromgetLaunchOptionsSync(), measures cold-start time (SDK init → first frame), and logsonShow/onHideforeground/background breadcrumbs.FrameRateIntegration(enableMinigameFrameRate) — samples the globalrequestAnimationFrameto estimate FPS and jank. Mini programs use a two-thread model with no logic-layerrequestAnimationFrame, so this safely no-ops there.
No extra wiring is needed — Sentry.init() auto-enables these in a minigame. To force them on or off:
Sentry.init({
dsn: '...',
enableMinigameLifecycle: true,
enableMinigameFrameRate: true,
});
For Each Agreed Feature
Walk through features one at a time. Load the corresponding reference file:
| Feature | Reference | Load when... |
|---|---|---|
| Error Monitoring | references/error-monitoring.md | Always |
| Performance Tracing | references/tracing.md | User agreed to tracing |
| Offline Cache | references/offline-cache.md | User agreed to offline cache |
| Source Map | references/sourcemap.md | User agreed to source maps |
Configuration Reference
Key Init Options
| Option | Type | Default | Description |
|---|---|---|---|
dsn | string | — | Sentry DSN (required) |
release | string | — | Release version, must match source map upload |
environment | string | — | Environment name (production, staging, etc.) |
sampleRate | number | 1.0 | Error event sample rate (0.0–1.0) |
tracesSampleRate | number | — | Trace sample rate (0.0–1.0) |
tracesSampler | function | — | Dynamic sampling function (overrides tracesSampleRate) |
enableSourceMap | boolean | true | Auto-normalize stack trace paths for source map resolution |
stackParser | StackParser | miniappStackParser | Custom stack parser for private runtimes or special stack formats |
enableOfflineCache | boolean | true | Cache events when offline, retry when back online |
offlineCacheLimit | number | 30 | Max events to store in offline cache |
offlineCacheMaxAge | number | 86400000 | Drop cached events older than this (ms); default 24h |
requireConsent | boolean | false | Gate outbound Sentry network sends until Sentry.setConsent(true) |
consentCacheLimit | number | 100 | Max events buffered before consent; preserves oldest cold-start data |
consentCacheMaxBytes | number | 921600 | Max consent-buffer bytes; default ~900KB due miniapp single-key storage limits |
consentCacheMaxAge | number | 86400000 | Drop consent-buffered events older than this (ms); default 24h |
onConsentCacheDrop | function | — | Called with { reason, dropped } when consent buffer drops events |
enableTracePropagation | boolean | true | Inject distributed tracing headers (sentry-trace/baggage, plus optional traceparent) in outgoing requests |
enableStandaloneHttpSpans | boolean | true | Send parentless API requests as standalone segment spans; set false for child-only request tracing |
tracePropagationTargets | Array | [] | URL allowlist for trace header injection; empty means no injection because mini programs have no reliable same-origin baseline |
propagateTraceparent | boolean | false | Also inject W3C traceparent for OpenTelemetry / W3C Trace Context compatible backends |
enableAutoSessionTracking | boolean | true | Automatic session lifecycle management |
enableConsoleBreadcrumbs | boolean | false | Capture console.log/warn/error as breadcrumbs |
enableLogs | boolean | false | Enable Sentry Logs via Sentry.logger.* |
traceNetworkBody | boolean | false | Capture request/response body in network breadcrumbs |
enableNavigationBreadcrumbs | boolean | true | Page lifecycle (navigation) breadcrumbs |
enableUserInteractionBreadcrumbs | boolean | true | Tap / user-interaction breadcrumbs |
enableNetworkStatusMonitoring | boolean | true | Real-time network monitoring; triggers offline flush on reconnect |
allowUrls | Array<string|RegExp> | — | Only send errors whose URL matches (others dropped) |
denyUrls | Array<string|RegExp> | — | Drop errors whose URL matches |
ignoreErrors | Array<string|RegExp> | — | Drop errors whose message/type matches |
transportOptions | object | see below | Built-in transport safeguards and headers |
defaultIntegrations | false|Integration[] | all built-in default integrations | Set false to disable every default integration; a custom array replaces the default base |
integrations | Integration[]|(defaults) => Integration[] | — | An array appends to defaults (user instances win by name); a function receives defaults and returns the final list |
enableMinigameLifecycle | boolean | minigame true / miniprogram false | Minigame cold-start + scene + show/hide breadcrumbs |
enableMinigameFrameRate | boolean | minigame true / miniprogram false | Minigame FPS / jank sampling (no-op in mini program) |
beforeSend | function | — | Event processor for filtering/modifying events |
beforeSendTransaction | function | — | Hook to filter/modify transaction events before sending |
beforeSendLog | function | — | Hook to filter/modify logs before sending |
beforeBreadcrumb | function | — | Hook to filter/modify breadcrumbs before they are attached |
The built-in transport defaults to requestTimeout: 3000 and maxConcurrentRequests: 2 so Sentry cannot occupy all mini program network slots when the service is unavailable. Additional envelopes wait in the bounded @sentry/core buffer. A timed-out request is aborted when the host returns an abortable request task, then handed to offline caching. Keep these defaults unless real-device testing shows a need to adjust them; transportOptions.headers remains available for custom envelope headers.
Privacy Consent Gate
For domestic mini program / mini game privacy flows, initialize with requireConsent: true. Before the user agrees, the SDK still collects errors, breadcrumbs, and performance data, but sends no Sentry network requests; events are buffered in miniapp storage and flushed after consent.
Sentry.init({
dsn: '...',
requireConsent: true,
consentCacheLimit: 100,
});
// After the user agrees to the privacy policy
Sentry.setConsent(true);
// If consent is revoked, block outbound sends again
Sentry.setConsent(false);
requireConsent implies local buffering even when enableOfflineCache is false; custom transport functions are wrapped by the consent gate too. The current store uses one storage key, so keep consentCacheMaxBytes near the default ~900KB unless the SDK adds sharded storage in a future version.
Platform Compatibility
| Feature | Alipay | ByteDance | Baidu | DingTalk | Kuaishou | ||
|---|---|---|---|---|---|---|---|
| Error Capture | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Performance | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Offline Cache | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Distributed Tracing | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Session Tracking | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Source Map | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
✅ = supported through the SDK's cross-platform abstraction, not a per-host on-device certification. WeChat / Alipay / ByteDance are the most battle-tested; validate less common hosts (Kuaishou, DingTalk) in your own environment before relying on them in production.
Verification
After setup, verify the integration:
- Trigger a test error:
// In any page
Sentry.captureException(new Error('Test error from mini program'));
-
Check Sentry dashboard:
- Go to your Sentry project → Issues
- You should see the test error with:
- Device info (brand, model, OS)
- Top-level
platform: javascript - Mini program context (host platform and host SDK version)
- App context (
app_versionfromminiProgram.versionwhen available) - Breadcrumbs (page navigation, network requests)
-
Verify automatic capture works:
- Throw an unhandled error in a page — it should appear in Sentry without any manual
captureExceptioncall
- Throw an unhandled error in a page — it should appear in Sentry without any manual
Phase 4: Cross-Link
After mini program setup, check for companion services:
# Check for backend services in the same workspace
ls -d ../*/package.json ../*/requirements.txt ../*/go.mod ../*/Gemfile 2>/dev/null
| Backend Detected | Suggest |
|---|---|
Node.js (package.json with server framework) | Sentry Node SDK |
Python (requirements.txt) | Sentry Python SDK |
Go (go.mod) | Sentry Go SDK |
Ruby (Gemfile) | Sentry Ruby SDK |
Java (pom.xml / build.gradle) | Sentry Java SDK |
Tip: Enable distributed tracing on both mini program and backend to get end-to-end request traces across services.
Troubleshooting
| Issue | Solution |
|---|---|
| Events not appearing in Sentry | Check DSN is correct; verify Sentry domain is in mini program's trusted domain list |
sampleRate filtering all events | Ensure sampleRate is not set to 0; default is 1.0 |
| Tracing spans not appearing | Set tracesSampleRate > 0 (or use tracesSampler) — tracing is off until set; there is no default |
| Minified stack traces | Set up Source Map upload — see references/sourcemap.md |
| Duplicate error reports | Do NOT manually call Sentry.captureException in onError — SDK captures automatically |
| Events lost on weak networks | Enable offline cache: enableOfflineCache: true (default) |
WeChat DevTools not triggering onError | Test on a real device; DevTools may not trigger all error handlers |
| Stack trace paths don't match source maps | Ensure --url-prefix "app:///" when uploading; SDK normalizes paths to app:/// automatically |
Frequently asked questions
What to verify before installation and use
What does the sentry-miniapp-sdk source document cover?
Set up Sentry error monitoring, performance tracing, and offline caching in mini program projects using sentry-miniapp — a community SDK built on @sentry/core v10.
How do I install sentry-miniapp-sdk?
The source record exposes this install command: npx skills add https://github.com/lizhiyao/sentry-miniapp --skill ".agents/skills/sentry-miniapp-sdk". Inspect the command and pinned source before running it.
Which permission-related actions were detected?
Static rules flagged exec-script, network, read-files in the source; the page lists the matching lines and excerpts.
Alternatives
Compare before choosing
coreyhaines31/marketingskills
ab-testing
When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program
garrytan/gbrain
bulk-ingestion
End-to-end discipline for turning any large data source (audio libraries, email takeouts, document corpora, chat exports, API dumps) into brain pages at scale. The lifecycle spine: SCHEMA → ACCESS → TRIAL → EVALUATE → IMPROVE → CODIFY → TEST → SKILLIFY → BULK → MONITOR. State is tracked in a durable JSON manifest (see MANIFEST-PATTERN.md) so any crash, session boundary, or subagent fan-out resumes from ground truth instead of memory.
alirezarezvani/claude-skills
app-store-optimization
App Store Optimization (ASO) toolkit for researching keywords, analyzing competitor rankings, generating metadata suggestions, and improving app visibility on Apple App Store and Google Play Store. Use when the user asks about ASO, app store rankings, app metadata, app titles and descriptions, app store listings, app visibility, or mobile app marketing on iOS or Android. Supports keyword research and scoring, competitor keyword analysis, metadata optimization, A/B test planning, launch checklist
dotnet/skills
migrate-vstest-to-mtp
Migrates .NET test projects from VSTest to Microsoft.Testing.Platform (MTP). Use when user asks to "migrate to MTP", "switch from VSTest", "enable Microsoft.Testing.Platform", "use MTP runner", set OutputType=Exe only for test projects in Directory.Build.props, or mentions EnableMSTestRunner, EnableNUnitRunner, or UseMicrosoftTestingPlatformRunner. USE FOR: MTP behavioral differences vs VSTest (exit code 8, zero tests discovered, --ignore-exit-code, TESTINGPLATFORM_EXITCODE_IGNORE); centralizing