sentry-javascript-bugs

작성자: sentry

Sentry React 및 TypeScript 변경 사항을 실제 프로덕션 이슈에서 추출된 버그 패턴을 기준으로 검토합니다. 프론트엔드 diff나 PR을 검토하거나 Warden을 확인할 때 사용하세요.

npx skills add https://github.com/getsentry/sentry --skill sentry-javascript-bugs

Sentry JavaScript Frontend Bug Pattern Review

Find bugs in Sentry frontend code by checking for the patterns that cause the most real production errors.

This skill encodes patterns from 428 real production issues (201 resolved, 130 ignored, 97 unresolved) generating over 524,000 error events across 93,000+ affected users. These are not theoretical risks -- they are the actual bugs that ship most often, with known fixes from resolved issues.

Scope

Review the code provided by the user, Warden, or the current branch diff. If the user does not provide a target, review the current branch diff. Start from the changed hunk or file, then read outward only as needed to confirm the behavior.

  1. Analyze the changed code against the pattern checks below.
  2. Use Read and Grep to trace data flow beyond the initial diff when needed. Follow component props, hook return values, API response shapes, and state transitions until the behavior is confirmed.
  3. Report only HIGH and MEDIUM confidence findings.
ConfidenceCriteriaAction
HIGHTraced the code path, confirmed the pattern matches a known bug classReport with fix
MEDIUMPattern is present but context may mitigate itReport as needs verification
LOWTheoretical or mitigated elsewhereDo not report

Step 1: Classify the Code

Determine what you are reviewing and load the relevant reference.

Code TypeLoad Reference
Null/undefined property access, optional chaining, object destructuringreferences/null-reference-errors.md
Dashboard widgets, chart visualization, widget URL generationreferences/dashboard-widget-errors.md
Trace views, span details, trace tree renderingreferences/trace-view-errors.md
API calls, response handling, error states, fetch wrappersreferences/api-response-handling.md
React hooks, context providers, render loops, component lifecyclereferences/react-lifecycle-errors.md
AI Insights, LLM prompt parsing, gen_ai span datareferences/ai-insights-parsing.md
Array operations, date/time values, numeric formattingreferences/range-and-bounds-errors.md

If the code spans multiple categories, load all relevant references.

Step 2: Check for Top Bug Patterns

These are ordered by combined frequency and impact from real production data.

Check 1: Null/Undefined Property Access -- 158 issues, 46,337 events

Code accesses a property on a value that may be null or undefined. This is the single most common bug pattern in the Sentry frontend.

Red flags:

  • Accessing .id, .slug, .name, .type, .match, .length, .charCodeAt without null checks
  • Using object.property instead of object?.property on data from API responses
  • Passing API response data directly to utility functions without null validation
  • Accessing DOM element properties from querySelector or useRef without checking if the element exists
  • Destructuring objects from hooks/stores that may return null during loading states
  • Calling .dispatchEvent() on elements that have been unmounted

Safe patterns:

  • Optional chaining: obj?.property?.nested
  • Default values: const value = obj?.field ?? defaultValue
  • Null guards before function calls: if (data) { parser.parse(data); }
  • Early returns for null/undefined parameters in utility functions

Check 2: Dashboard Widget Input Validation -- 6 issues, 90,482 events

Widget visualization components throw when receiving data in unexpected formats.

Red flags:

  • Rendering chart components without checking if data contains plottable values
  • Calling getWidgetExploreUrl() for widget types that do not support multiple queries
  • Passing undefined field values to parseFunction() or similar field parsers
  • Not handling empty API responses in widget data fetchers

Safe patterns:

  • Validate data shape before rendering: if (!hasPlottableValues(data)) return <EmptyState />
  • Check widget query count before generating explore URLs
  • Guard field parsers: if (!field) return null

Check 3: Trace View Data Integrity -- 12 issues, 328,482 events

The trace tree renderer and trace detail views encounter data that violates structural assumptions.

Red flags:

  • Building trace trees without cycle detection (or detecting cycles but not handling them gracefully)
  • Looking up projects by ID from span data without checking if the project is accessible
  • Generating trace links without validating traceSlug is non-empty
  • Using captureException in render paths without deduplication (fires every render cycle)

Safe patterns:

  • Break cycles by detaching cyclic nodes as orphan roots
  • Validate traceSlug before generating links: if (!traceSlug) return fallbackLink
  • Deduplicate error captures using a ref: if (!capturedRef.current) { captureException(...); capturedRef.current = true; }
  • Check project access before rendering span details

Check 4: API Response Shape Assumptions -- 31 issues, 24,019 events

Frontend code assumes API responses have a specific shape but the response is empty, undefined, or has an unexpected status code.

Red flags:

  • Not handling 200 responses with empty bodies (e.g., GET /customers/{orgSlug}/ returns 200 with no body)
  • Not handling 402 (Payment Required) status codes in subscription flows
  • Not handling 409 (Conflict) status codes in mutation endpoints
  • Treating UndefinedResponseBodyError as unexpected (it indicates the API returned no parseable body)
  • Assuming SelectAsync options will always load successfully

Safe patterns:

  • Check response body before parsing: if (!response.body) return null
  • Handle specific 4xx status codes in catch blocks
  • Provide fallback empty states for failed API fetches instead of throwing

Check 5: React Lifecycle Violations -- 10 issues, 2,595 events

Components violate React rendering rules, causing infinite loops or crashes.

Red flags:

  • Setting state unconditionally in useEffect without proper dependency arrays
  • Calling useOrganization() in components that render before organization context is loaded
  • Using useContext() outside the provider boundary
  • Passing objects as React children instead of strings/elements
  • Components that trigger immediate re-render on mount

Safe patterns:

  • Always provide dependency arrays for useEffect
  • Guard context hooks: const org = useOrganization(); if (!org) return <Loading />
  • Wrap organization-dependent routes in a provider boundary
  • Validate element types before rendering: if (typeof Component !== 'function') return null

Check 6: AI Insights Data Parsing -- 2 issues, 3,005 events

JSON parsing of AI prompt messages and gen_ai span data fails on non-standard formats.

Red flags:

  • Calling JSON.parse() on ai.prompt.messages span attributes without try-catch
  • Assuming all AI model responses produce valid JSON
  • Not handling the "parts" format for multi-modal AI messages

Safe patterns:

  • Wrap all JSON.parse calls on external data in try-catch
  • Check for leading [ or { before parsing
  • Provide raw-text fallback rendering when parsing fails

Check 7: Array and Bounds Validation -- 15 issues, 3,120 events

Array operations and numeric formatting with values that exceed valid ranges.

Red flags:

  • Using result.push(...largeArray) (crashes when array is too large)
  • Passing unclamped values to toLocaleString({maximumFractionDigits: n})
  • Constructing Date objects from unvalidated timestamps
  • Recursive component rendering without depth limits

Safe patterns:

  • Use concat or iterative push for potentially large arrays
  • Clamp numeric format parameters: Math.min(100, Math.max(0, precision))
  • Validate dates before constructing: if (isNaN(new Date(ts).getTime())) return fallback
  • Use iterative rendering with explicit stacks for deeply nested structures

Check 8: Logic Correctness -- not pattern-based

After checking all known patterns above, reason about the changed code itself:

  • Does every code path return the correct type (or JSX)?
  • Are all branches of conditionals handled (especially missing else / default cases in switches)?
  • Can any prop or state value (null, undefined, empty array, empty string) cause unexpected behavior?
  • Are hook dependency arrays correct? Missing deps cause stale closures; extra deps cause infinite loops.
  • If this component unmounts mid-async-operation, is cleanup handled?

Only report if you can trace a specific input that triggers the bug. Do not report theoretical concerns.

If no checks produced a potential finding, stop and report zero findings. Do not invent issues to fill the report. An empty result is the correct output when the code has no bugs matching these patterns.

Each code location should be reported once under the most specific matching pattern. Do not flag the same line under multiple checks.

Step 3: Report Findings

For each finding, provide the evidence the review harness needs:

  • precise location
  • severity and confidence
  • concrete triggering input or state
  • root cause and consequence
  • a matching production precedent when available
  • a concrete code fix, preferably as a unified diff when the harness supports it

Fix suggestions must include actual code. Never suggest a comment or docstring as a fix.

Do not prescribe your own output format — the review harness controls the response structure.

sentry의 다른 스킬

generate-frontend-forms
sentry
Sentry의 새로운 폼 시스템을 사용하여 폼을 생성하는 가이드입니다. 폼, 폼 필드, 유효성 검사 또는 자동 저장 기능을 구현할 때 사용하세요.
official
sentry-snapshots-cocoa
sentry
Apple/Cocoa 프로젝트를 위한 전체 Sentry Snapshots 설정입니다. "SnapshotPreviews 설정", "Apple 스냅샷 테스트 설정", "Apple 스냅샷 업로드" 요청 시 사용하세요.
official
architecture-review
sentry
직원 수준의 코드베이스 건강 검토. 모놀리식 모듈, 무음 실패, 타입 안전성 격차, 테스트 커버리지 구멍, LLM 친화성 문제를 찾습니다.
official
linear-type-labeler
sentry
Linear 이슈를 분류하고, 각 이슈의 제목과 설명 내용을 기반으로 Sentry 워크스페이스의 레이블 분류 체계에서 Type 레이블을 적용합니다.
official
sentry-flutter-sdk
sentry
Flutter 및 Dart를 위한 완전한 Sentry SDK 설정입니다. "Flutter에 Sentry 추가", "sentry_flutter 설치", "Dart에서 Sentry 설정" 또는 오류 구성을 요청받았을 때 사용하세요.
official
sentry-svelte-sdk
sentry
Svelte 및 SvelteKit을 위한 완전한 Sentry SDK 설정입니다. "Svelte에 Sentry 추가", "SvelteKit에 Sentry 추가", "@sentry/sveltekit 설치" 또는 구성 요청 시 사용하세요.
official
vercel-react-best-practices
sentry
Vercel Engineering의 React 및 Next.js 성능 최적화 가이드라인입니다. 이 스킬은 React/Next.js 코드를 작성, 검토 또는 리팩토링할 때 사용해야 합니다.
official
sentry-tanstack-start-sdk
sentry
TanStack Start React용 전체 Sentry SDK 설정. "TanStack Start에 Sentry 추가", "@sentry/tanstackstart-react 설치" 또는 오류 구성 요청 시 사용…
official