skills/sentry-tanstack-start-sdk/SKILL.md
Full Sentry SDK setup for TanStack Start React. Use when asked to "add Sentry to TanStack Start", "install @sentry/tanstackstart-react", or configure error monitoring, tracing, session replay, logs, or user feedback in a TanStack Start React app.
npx skillsauth add getsentry/sentry-for-claude sentry-tanstack-start-sdkInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
All Skills > SDK Setup > TanStack Start React SDK
Opinionated wizard that scans your TanStack Start React project and guides you through complete Sentry setup for browser and server runtimes.
@sentry/tanstackstart-reactsentryTanstackStart, wrapFetchWithSentry, instrument.server.mjs, or TanStack Start middleware instrumentationNote: This SDK is currently alpha and documented as compatible with TanStack Start
1.0 RC. Always verify against docs.sentry.io/platforms/javascript/guides/tanstackstart-react/ before implementing.
Run these commands to understand the project before making any recommendations:
# Detect TanStack Start / Router and existing Sentry
cat package.json | grep -E '"@tanstack/react-start"|"@tanstack/react-router"|"@sentry/tanstackstart-react"'
# Check if Sentry is already present
cat package.json | grep '"@sentry/'
# Detect key files used by the TanStack Start setup
ls src/router.tsx src/start.ts src/server.ts instrument.server.mjs vite.config.ts vite.config.js 2>/dev/null
# Check whether source map upload credentials are configured
cat .env .env.local .env.sentry-build-plugin 2>/dev/null | grep "SENTRY_AUTH_TOKEN"
# Detect deployment hints in scripts
cat package.json | grep -E '"dev"|"build"|"start"|NODE_OPTIONS|--import'
# Detect logging libraries
cat package.json | grep -E '"pino"|"winston"|"loglevel"'
# Detect companion backend directories
ls ../backend ../server ../api 2>/dev/null
cat ../go.mod ../requirements.txt ../Gemfile ../pom.xml 2>/dev/null | head -3
What to determine:
| Question | Impact |
|----------|--------|
| @tanstack/react-start present? | Confirms this skill is the right setup path |
| @sentry/tanstackstart-react already installed? | Skip install and go to feature tuning |
| src/router.tsx exists? | Client-side Sentry.init placement |
| src/start.ts exists? | Global middleware setup for server-side errors |
| src/server.ts exists? | Server entry instrumentation placement |
| instrument.server.mjs exists? | Runtime startup instrumentation path |
| vite.config.ts exists? | Add sentryTanstackStart plugin and source maps |
| SENTRY_AUTH_TOKEN configured? | Source map upload readiness |
| Backend directory found? | Trigger Phase 4 cross-link suggestion |
Present a concrete recommendation based on what you found. Do not ask open-ended questions — lead with a proposal:
Recommended (core coverage):
Optional (enhanced observability):
Recommendation logic:
| Feature | Recommend when... | |---------|------------------| | Error Monitoring | Always — non-negotiable baseline | | Tracing | Usually yes for TanStack Start; route + fetch instrumentation gives immediate value | | Session Replay | User-facing app, login flows, checkout flows, or hard-to-reproduce UX bugs | | Logs | Existing logging strategy, support workflow, or trace/log correlation needs | | User Feedback | Team wants direct user reports without leaving the app |
Propose: "I recommend Error Monitoring + Tracing + Session Replay. Want me to also enable Logs and User Feedback?"
npm install @sentry/tanstackstart-react --save
src/router.tsxInitialize Sentry inside the router factory and gate it to the browser:
import * as Sentry from "@sentry/tanstackstart-react";
import { createRouter } from "@tanstack/react-router";
export const getRouter = () => {
const router = createRouter();
if (!router.isServer) {
Sentry.init({
dsn: "___PUBLIC_DSN___",
sendDefaultPii: true,
integrations: [
Sentry.tanstackRouterBrowserTracingIntegration(router),
Sentry.replayIntegration(),
Sentry.feedbackIntegration({
colorScheme: "system",
}),
],
enableLogs: true,
tracesSampleRate: 1.0,
replaysSessionSampleRate: 0.1,
replaysOnErrorSampleRate: 1.0,
});
}
return router;
};
instrument.server.mjsCreate instrument.server.mjs in project root:
import * as Sentry from "@sentry/tanstackstart-react";
Sentry.init({
dsn: "___PUBLIC_DSN___",
sendDefaultPii: true,
enableLogs: true,
tracesSampleRate: 1.0,
});
vite.config.tssentryTanstackStart should be the last plugin:
import { defineConfig } from "vite";
import { sentryTanstackStart } from "@sentry/tanstackstart-react/vite";
import { tanstackStart } from "@tanstack/react-start/plugin/vite";
export default defineConfig({
plugins: [
tanstackStart(),
sentryTanstackStart({
org: "___ORG_SLUG___",
project: "___PROJECT_SLUG___",
authToken: process.env.SENTRY_AUTH_TOKEN,
}),
],
});
If the token is stored in .env, load it with loadEnv in the Vite config before passing it to the plugin.
src/server.tsWrap the fetch handler with wrapFetchWithSentry:
import { wrapFetchWithSentry } from "@sentry/tanstackstart-react";
import handler, { createServerEntry } from "@tanstack/react-start/server-entry";
export default createServerEntry(
wrapFetchWithSentry({
fetch(request: Request) {
return handler.fetch(request);
},
}),
);
src/start.tsThese middleware capture server-side request and function errors:
import {
sentryGlobalFunctionMiddleware,
sentryGlobalRequestMiddleware,
} from "@sentry/tanstackstart-react";
import { createStart } from "@tanstack/react-start";
export const startInstance = createStart(() => {
return {
requestMiddleware: [sentryGlobalRequestMiddleware],
functionMiddleware: [sentryGlobalFunctionMiddleware],
};
});
Sentry middleware should be first in each array.
Choose one runtime method:
| Runtime pattern | Use when... | Notes |
|---|---|---|
| --import flag | You can control Node startup flags | Preferred for production monitoring |
| Direct import in src/server.ts | Host restricts startup flags (for example serverless hosts) | Limits instrumentation to native Node APIs |
--import examples:
{
"scripts": {
"dev": "NODE_OPTIONS='--import ./instrument.server.mjs' vite dev --port 3000",
"build": "vite build && cp instrument.server.mjs .output/server",
"start": "node --import ./.output/server/instrument.server.mjs .output/server/index.mjs"
}
}
Direct import fallback (top of src/server.ts):
import "../instrument.server.mjs";
Walk through features one at a time. Load the reference file, follow steps exactly, and verify before moving on:
| Feature | Reference | Load when... |
|---------|-----------|-------------|
| Error Monitoring | ${SKILL_ROOT}/references/error-monitoring.md | Always |
| Tracing | ${SKILL_ROOT}/references/tracing.md | Route/API performance visibility needed |
| Session Replay | ${SKILL_ROOT}/references/session-replay.md | User-facing app |
| Logs | ${SKILL_ROOT}/references/logging.md | Structured logs and correlation needed |
| User Feedback | ${SKILL_ROOT}/references/user-feedback.md | In-app feedback collection needed |
| TanStack Start Features | ${SKILL_ROOT}/references/tanstackstart-features.md | Server entry, Vite plugin, source maps, runtime startup |
For each feature: Read ${SKILL_ROOT}/references/<feature>.md, follow steps exactly, verify it works.
Sentry.init() Options| Option | Type | Default | Notes |
|--------|------|---------|-------|
| dsn | string | — | Required; SDK is disabled when empty |
| sendDefaultPii | boolean | false | Sends request headers and IP-derived user context |
| integrations | Integration[] | SDK defaults | Include TanStack Router tracing, replay, feedback as needed |
| enableLogs | boolean | false | Enables Sentry.logger.* APIs |
| tracesSampleRate | number | — | 1.0 in development, lower in production |
| replaysSessionSampleRate | number | — | Fraction of all sessions recorded |
| replaysOnErrorSampleRate | number | — | Fraction of error sessions recorded |
| tunnel | string | — | Optional ad-blocker bypass endpoint |
| debug | boolean | false | SDK diagnostic logging |
| API | Purpose |
|-----|---------|
| tanstackRouterBrowserTracingIntegration(router) | Browser navigation tracing |
| wrapFetchWithSentry(...) | Server request tracing + error capture on fetch handler |
| sentryGlobalRequestMiddleware | Captures request-level server errors |
| sentryGlobalFunctionMiddleware | Captures server function errors |
| sentryTanstackStart({...}) | Vite plugin for source maps and middleware instrumentation |
Trigger test events to confirm Sentry receives data.
<button
type="button"
onClick={() => {
throw new Error("Sentry Test Error");
}}
>
Break the world
</button>
<button
type="button"
onClick={async () => {
await Sentry.startSpan({ name: "Example Frontend Span", op: "test" }, async () => {
const res = await fetch("/api/sentry-example");
if (!res.ok) {
throw new Error("Sentry Example Frontend Error");
}
});
}}
>
Break the world
</button>
Sentry.logger.info("User example action completed");
Sentry.logger.warn("Slow operation detected", { operation: "data_fetch", duration: 3500 });
Sentry.logger.error("Validation failed", { field: "email", reason: "Invalid email" });
Confirm in Sentry:
enableLogs: trueAfter completing TanStack Start setup, check whether a companion backend exists without Sentry:
ls ../backend ../server ../api ../go ../python 2>/dev/null
cat ../go.mod ../requirements.txt ../pyproject.toml ../Gemfile ../pom.xml 2>/dev/null | head -5
If a backend exists without Sentry configured, suggest the matching skill:
| Backend detected | Suggest skill |
|------------------|--------------|
| Go (go.mod) | sentry-go-sdk |
| Python (requirements.txt, pyproject.toml) | sentry-python-sdk |
| Ruby (Gemfile) | sentry-ruby-sdk |
| Java (pom.xml, build.gradle) | Use @sentry/java docs |
| Node.js backend services | sentry-node-sdk |
| Issue | Solution |
|-------|----------|
| Events not appearing | Set debug: true, verify DSN, and ensure client/server init files both run |
| No server traces | Confirm src/server.ts uses wrapFetchWithSentry and runtime loads instrument.server.mjs |
| Server errors missing from route handlers | Ensure sentryGlobalRequestMiddleware and sentryGlobalFunctionMiddleware are first in arrays |
| Source maps not resolving | Verify SENTRY_AUTH_TOKEN, org, and project in sentryTanstackStart config |
| SENTRY_AUTH_TOKEN undefined in Vite config | Use loadEnv(mode, process.cwd(), "") or .env.sentry-build-plugin |
| Replay not recording | Ensure replayIntegration() is in integrations and sample rates are non-zero |
| Feedback widget not visible | Confirm feedbackIntegration() is configured and check CSS z-index conflicts |
| Logs missing in Sentry | Set enableLogs: true and use Sentry.logger.* APIs |
| Direct-import setup misses library spans | Prefer --import startup when possible; direct import supports native Node instrumentation only |
| SSR rendering exceptions not auto-captured | Capture manually with Sentry.captureException in error boundaries / fallback handlers |
development
Decide which Sentry signal to reach for when instrumenting code — error, span, span attribute, log, or metric. Use when adding instrumentation and unsure whether something should be a log vs a span vs a metric, when deciding "what to instrument where", when reviewing instrumentation for gaps, or when a coding agent needs a rule for choosing between errors, traces, logs, and metrics. This skill decides WHAT to emit; the sentry-*-sdk skills handle HOW to set each pillar up.
development
Full Sentry SDK setup for React Native and Expo. Use when asked to "add Sentry to React Native", "install @sentry/react-native", "setup Sentry in Expo", or configure error monitoring, tracing, profiling, session replay, or logging for React Native applications. Supports Expo managed, Expo bare, and vanilla React Native.
development
Full Sentry SDK setup for Python. Use when asked to "add Sentry to Python", "install sentry-sdk", "setup Sentry in Python", or configure error monitoring, tracing, profiling, logging, metrics, crons, or AI monitoring for Python applications. Supports Django, Flask, FastAPI, Celery, Starlette, AIOHTTP, Tornado, and more.
development
Configure specific Sentry features beyond basic SDK setup. Use when asked to monitor AI/LLM calls, set up OpenTelemetry pipelines, or create alerts and notifications.