Conversation
|
Warning Rate limit exceeded
⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 Walkthrough워크스루OpenAI 및 GitHub API를 이용해 PR의 변경(diff)을 분석하고 구조화된 요약(제목, 레이블, 본문)을 생성해 PR을 자동으로 업데이트하는 기능과 이를 위한 설정·유틸·스크립트·테스트·워크플로우를 추가합니다. 변경사항
시퀀스 다이어그램sequenceDiagram
participant GH as GitHub API
participant Script as PR AI Script
participant LM as LangSmith
participant AI as OpenAI API
participant Git as Git (Fallback)
Script->>GH: 레이블 목록 조회
GH-->>Script: 레이블 반환
Script->>GH: PR 정보 조회 (base/head SHA)
GH-->>Script: PR 정보
rect rgba(100, 149, 237, 0.5)
Note over Script,Git: Diff 수집
Script->>GH: Compare API로 diff 요청
alt API 성공
GH-->>Script: Diff 데이터
else API 실패 또는 제외조건 충족
Script->>Git: 로컬 git diff 생성
Git-->>Script: Diff 데이터
end
end
rect rgba(144, 238, 144, 0.5)
Note over Script: Diff 전처리
Script->>Script: 민감정보 마스킹
Script->>Script: 파일 필터링 및 트런케이션
end
Script->>Script: AI 프롬프트 구성
Script->>LM: 추적 기록 시작
Script->>AI: Chat Completion 요청 (JSON 스키마 응답)
AI-->>Script: 구조화된 요약 응답
Script->>LM: 추적 기록 완료
Script->>Script: 응답 스키마 검증
rect rgba(255, 182, 193, 0.5)
Note over Script,GH: PR 업데이트 적용
Script->>GH: 제목/본문 패치
GH-->>Script: 확인
Script->>GH: 담당자 할당 / 레이블 추가
GH-->>Script: 확인
end
예상 코드 리뷰 소요시간🎯 4 (복잡) | ⏱️ ~55분 🚥 Pre-merge checks | ✅ 1 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (1 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
github-actions
bot
changed the title
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/scripts/pr-ai-description-lib.mjs:
- Around line 401-416: maskSensitiveContent 함수의 현재 정규식이 제한적이어서 OPENAI_API_KEY,
DATABASE_URL, client_secret, AWS_SECRET_ACCESS_KEY 같은 흔한 env/키/URL 패턴을 마스킹하지
못합니다; maskSensitiveContent를 수정해 env-var 스타일 키(예:
/[A-Z0-9_]*(_KEY|_SECRET|_TOKEN|_PASSWORD|API_KEY|ACCESS_KEY|CLIENT_SECRET)/i),
키=값 및 키:값 할당 패턴, URL에 포함된 사용자정보(예: https?://user:pass@host)와 .env 형식(예:
/^(\w+)=.+$/m) 등을 추가로 검사해 값 부분을 '[REDACTED]'로 바꾸고, 기존
Authorization/Bearer/password/secret/token/apiKey 규칙과 결합하여 포괄적으로 처리하세요; 또한
민감파일(.env, .env.*, *credentials*, key files)은
.github/scripts/pr-ai-description.mjs 및 .github/scripts/pr-ai-dry-run.mjs에서 전송 전
diff 대상에서 제외하도록 배포 스크립트 또는 업로드 로직에서 파일 이름 필터링을 추가하세요.
- Around line 556-558: The PR table breaks when change.description contains line
breaks; before escaping pipes, normalize newlines in summary.changes entries
(e.g., in the loop that creates escapedDesc) by replacing \r\n, \r, and \n with
a single '<br>' or a space, then escape '|' and push the row; update the logic
around summary.changes / change.description and the escapedDesc construction
used in lines.push so all newlines are normalized prior to rendering the
Markdown table (validateAiSummaryJson() can still allow multiline input).
In @.github/scripts/pr-ai-description.mjs:
- Around line 358-375: The current fetch with '--depth=1' then running git diff
between baseSha and headSha can fail in shallow checkouts because merge-base may
be missing; update the logic around runGitCommand/fetch and the subsequent
range/nameStatus steps to detect when git merge-base (or git diff) fails and, as
a fallback, deepen the shallow clone (e.g. retry runGitCommand(['fetch',
'--deepen', <N>, 'origin']) or re-fetch without depth) in a loop until git
merge-base baseSha headSha succeeds or a max depth/attempts is reached, then
recompute range and call runGitCommand([... 'diff' ...]) so nameStatus is
produced from a repo that contains the merge base.
In @.github/scripts/pr-ai-dry-run.mjs:
- Around line 229-233: Wrap the call that builds the PR stub's user login (the
runGit(['config', 'user.name']) invocation inside the pr object) in a try/catch
(or otherwise call git config --get user.name and handle non-zero exit) so a
missing git user doesn't cause the script to exit; if the git call throws or
returns nothing, set user.login to 'local' (i.e., ensure the construction of
pr.user uses a safe getter with fallback to 'local' instead of relying on ||
after a failing runGit call).
- Around line 153-173: runWithPr() currently passes the raw compare result
straight to normalizeDiffEntries(), bypassing the same missing-patch and
maxFiles checks used in the real workflow; update runWithPr() to reuse the
shared diff-collection flow: call fetchCompareDiff(baseSha, headSha), then run
buildExcludeGlobs() -> filterDiffFiles() and perform the same "no included
files" / "missing patch" / "too many files" checks and git-fallback or abort
logic used by the production diff collector before calling
normalizeDiffEntries(), so generateAndPrint() receives the same diffEntries as
the real workflow; adjust fetchCompareDiff(), filterDiffFiles(), or expose a
single collectDiffEntries() helper if needed to avoid duplicating logic.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: b3f0b9bd-434a-42d0-8243-0d968b3f2614
⛔ Files ignored due to path filters (1)
yarn.lockis excluded by!**/yarn.lock,!**/*.lock
📒 Files selected for processing (8)
.coderabbit.yaml.github/scripts/__tests__/pr-ai-description-lib.spec.mjs.github/scripts/__tests__/pr-ai-description.spec.mjs.github/scripts/pr-ai-description-lib.mjs.github/scripts/pr-ai-description.mjs.github/scripts/pr-ai-dry-run.mjs.github/workflows/pr-ai-description.ymlpackage.json
Comment on lines
+401
to
+416
| export function maskSensitiveContent(text) { | ||
| if (typeof text !== 'string' || text.length === 0) { | ||
| return ''; | ||
| } | ||
|
|
||
| return text | ||
| .replaceAll(/(Authorization\s*[:=]\s*)([^\n\r]+)/gi, '$1[REDACTED]') | ||
| .replaceAll(/Bearer\s+[A-Za-z0-9\-._~+/]+=*/gi, 'Bearer [REDACTED]') | ||
| .replaceAll( | ||
| /(\"?(?:password|secret|token|apiKey)\"?\s*[:=]\s*)\"([^\"\n\r]*)\"/gi, | ||
| '$1"[REDACTED]"', | ||
| ) | ||
| .replaceAll( | ||
| /(\"?(?:password|secret|token|apiKey)\"?\s*[:=]\s*)([^\s,\n\r]+)/gi, | ||
| '$1[REDACTED]', | ||
| ); |
There was a problem hiding this comment.
현재 마스킹 규칙으로는 비밀값이 그대로 외부 서비스로 전송될 수 있습니다.
지금 정규식은 Authorization, Bearer, 그리고 키 이름이 정확히 password|secret|token|apiKey인 경우만 가립니다. 그래서 OPENAI_API_KEY=..., DATABASE_URL=..., client_secret=..., AWS_SECRET_ACCESS_KEY=... 같은 흔한 형태는 그대로 남고, 이 값들은 이후 .github/scripts/pr-ai-description.mjs와 .github/scripts/pr-ai-dry-run.mjs에서 OpenAI/LangSmith로 전송됩니다. env-var/URL/credential 파일 패턴까지 마스킹 범위를 넓히고, .env*나 키 파일은 아예 diff 대상에서 제외하는 방어선도 같이 두는 편이 안전합니다.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.github/scripts/pr-ai-description-lib.mjs around lines 401 - 416,
maskSensitiveContent 함수의 현재 정규식이 제한적이어서 OPENAI_API_KEY, DATABASE_URL,
client_secret, AWS_SECRET_ACCESS_KEY 같은 흔한 env/키/URL 패턴을 마스킹하지 못합니다;
maskSensitiveContent를 수정해 env-var 스타일 키(예:
/[A-Z0-9_]*(_KEY|_SECRET|_TOKEN|_PASSWORD|API_KEY|ACCESS_KEY|CLIENT_SECRET)/i),
키=값 및 키:값 할당 패턴, URL에 포함된 사용자정보(예: https?://user:pass@host)와 .env 형식(예:
/^(\w+)=.+$/m) 등을 추가로 검사해 값 부분을 '[REDACTED]'로 바꾸고, 기존
Authorization/Bearer/password/secret/token/apiKey 규칙과 결합하여 포괄적으로 처리하세요; 또한
민감파일(.env, .env.*, *credentials*, key files)은
.github/scripts/pr-ai-description.mjs 및 .github/scripts/pr-ai-dry-run.mjs에서 전송 전
diff 대상에서 제외하도록 배포 스크립트 또는 업로드 로직에서 파일 이름 필터링을 추가하세요.
Member
Author
There was a problem hiding this comment.
@coderabbitai
diff에 OPENAI_API_KEY 실제값 같은 게 나오려면 누군가 시크릿을 코드에 하드코딩해서 커밋해야 함.
.env는 이미 gitignore 대상이고, 이 함수는 만약 남아있을 경우의 2차 방어선이어야 함.
현재 패턴으로 실무적으로 충분하다고 생각하는데 어떻게 생각함?
There was a problem hiding this comment.
@chanwoo7 맞습니다. 지적하신 점이 타당합니다.
말씀대로 실제 비밀값이 diff에 등장하려면 누군가가 코드에 직접 하드코딩해서 커밋해야 하고, .env는 gitignore로 이미 배제되어 있으니 이 함수가 처리해야 하는 현실적인 케이스는 상당히 좁습니다. 현재 패턴(Authorization, Bearer, password/secret/token/apiKey 키워드)은 코드 상에서 가장 흔히 실수로 커밋되는 형태를 잘 커버하고 있습니다.
굳이 보완이 필요한 edge case를 꼽자면:
- DB 연결 문자열(
postgresql://user:pass@host) 같은 URL 내 userinfo .env.example등 실값을 실수로 채운 채 커밋된 파일
하지만 이것도 2차 방어선의 범위를 어디까지 잡느냐의 설계 판단이고, 현재 구현으로도 실무적으로 합리적이라는 의견에 동의합니다. 현재 패턴을 유지하겠습니다.
✏️ Learnings added
Learnt from: chanwoo7
Repo: CaQuick/caquick-be PR: 9
File: .github/scripts/pr-ai-description-lib.mjs:409-424
Timestamp: 2026-03-15T16:10:56.277Z
Learning: In .github/scripts/pr-ai-description-lib.mjs, the maskSensitiveContent function is intentionally designed as a secondary defense layer only. The primary protection is .gitignore (which excludes .env files). The current regex patterns covering Authorization, Bearer, and password/secret/token/apiKey keywords are considered sufficient by the maintainer for this secondary role. Do not flag missing env-var style patterns (e.g., OPENAI_API_KEY, DATABASE_URL) as a critical issue.
Note: Learnings are effective only in the context of similar code segments. To apply general coding guidelines, please configure review instructions. You can manage existing learnings in the app.
Comment on lines
+358
to
+375
| runGitCommand([ | ||
| 'fetch', | ||
| '--no-tags', | ||
| '--prune', | ||
| '--depth=1', | ||
| 'origin', | ||
| baseSha, | ||
| headSha, | ||
| ]); | ||
|
|
||
| const range = `${baseSha}...${headSha}`; | ||
| const nameStatus = runGitCommand([ | ||
| 'diff', | ||
| '--no-color', | ||
| '--diff-algorithm=histogram', | ||
| '--name-status', | ||
| range, | ||
| ]); |
There was a problem hiding this comment.
git fallback이 shallow history에서는 복구 경로가 되지 못합니다.
여기서는 --depth=1로 baseSha/headSha만 가져온 뒤 git diff ${baseSha}...${headSha}를 실행합니다. 현재 워크플로도 shallow checkout이라 merge-base가 히스토리에 없는 경우가 많아서, compare API가 빠진 PR에서 오히려 no merge base로 다시 실패할 수 있습니다. fallback 경로라면 merge-base를 계산할 수 있을 만큼 히스토리를 더 가져오거나 git merge-base가 성공할 때까지 deepen해야 합니다.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.github/scripts/pr-ai-description.mjs around lines 358 - 375, The current
fetch with '--depth=1' then running git diff between baseSha and headSha can
fail in shallow checkouts because merge-base may be missing; update the logic
around runGitCommand/fetch and the subsequent range/nameStatus steps to detect
when git merge-base (or git diff) fails and, as a fallback, deepen the shallow
clone (e.g. retry runGitCommand(['fetch', '--deepen', <N>, 'origin']) or
re-fetch without depth) in a loop until git merge-base baseSha headSha succeeds
or a max depth/attempts is reached, then recompute range and call
runGitCommand([... 'diff' ...]) so nameStatus is produced from a repo that
contains the merge base.
Comment on lines
+153
to
+173
| log(`diff 수집 중... (${baseSha.slice(0, 8)}...${headSha.slice(0, 8)})`); | ||
| const compareFiles = fetchCompareDiff(baseSha, headSha); | ||
|
|
||
| const excludeGlobs = buildExcludeGlobs(); | ||
| const { included } = filterDiffFiles(compareFiles, excludeGlobs); | ||
|
|
||
| if (included.length === 0) { | ||
| throw new Error('필터링 후 diff 파일이 없습니다.'); | ||
| } | ||
|
|
||
| const diffEntries = normalizeDiffEntries(included); | ||
| const repositoryLabels = fetchRepositoryLabels(); | ||
|
|
||
| return generateAndPrint({ | ||
| pr, | ||
| diffEntries, | ||
| diffSource: 'compare', | ||
|
|
||
| repositoryLabels, | ||
| ...options, | ||
| }); |
There was a problem hiding this comment.
PR 모드 dry-run이 실제 워크플로와 다른 diff를 요약합니다.
runWithPr()는 compare 결과를 바로 normalizeDiffEntries()로 넘겨서 patch 누락 감지와 maxFiles 상한을 모두 건너뜁니다. 실제 워크플로의 .github/scripts/pr-ai-description.mjs는 compare missing patch / too many files를 감지한 뒤 git fallback 또는 중단을 결정하므로, large diff나 binary 파일이 섞인 PR에서 dry-run만 불완전한 요약을 만들 수 있습니다. 공통 diff 수집 로직으로 묶지 않으면 두 경로가 계속 어긋납니다.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.github/scripts/pr-ai-dry-run.mjs around lines 153 - 173, runWithPr()
currently passes the raw compare result straight to normalizeDiffEntries(),
bypassing the same missing-patch and maxFiles checks used in the real workflow;
update runWithPr() to reuse the shared diff-collection flow: call
fetchCompareDiff(baseSha, headSha), then run buildExcludeGlobs() ->
filterDiffFiles() and perform the same "no included files" / "missing patch" /
"too many files" checks and git-fallback or abort logic used by the production
diff collector before calling normalizeDiffEntries(), so generateAndPrint()
receives the same diffEntries as the real workflow; adjust fetchCompareDiff(),
filterDiffFiles(), or expose a single collectDiffEntries() helper if needed to
avoid duplicating logic.
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (3)
.github/scripts/pr-ai-description-lib.mjs (3)
6-14:⚠️ Potential issue | 🔴 Critical민감 파일 패턴이 기본 제외 목록에서 빠져 있습니다.
Line 6-14의 기본 제외 목록만으로는
.env, 키 파일, credential 파일이 diff에 포함되어 외부 서비스로 전달될 수 있습니다.수정 예시
export const DEFAULT_EXCLUDE_GLOBS = [ + '.env', + '.env.*', + '**/*credentials*', + '**/*.pem', + '**/*.key', 'yarn.lock', 'package-lock.json', 'pnpm-lock.yaml', '**/*.snap', 'dist/**', 'coverage/**', '**/*.map', ];🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/pr-ai-description-lib.mjs around lines 6 - 14, The DEFAULT_EXCLUDE_GLOBS array currently omits common sensitive files so secrets like .env, key/cert files, and credential stores may be included in diffs; update the DEFAULT_EXCLUDE_GLOBS constant to add patterns that exclude environment and credential files (e.g., .env, .env.*, *.key, *.pem, *.p12, *.jks, *.crt, *.der, *.secret, credentials files and common credential paths like .aws/credentials and .git-credentials) so these sensitive patterns are never exported by the PR AI description script.
556-558:⚠️ Potential issue | 🟡 Minor
changes.description개행 미정규화로 Markdown 표가 깨질 수 있습니다.Line 557은
|만 escape하고 있어서\r/\n이 들어오면 row가 분리됩니다. 표 렌더링 안정성을 위해 개행을 먼저 정규화해 주세요.수정 예시
for (const change of summary.changes) { - const escapedDesc = change.description.replaceAll('|', '\\|'); + const escapedDesc = change.description + .replaceAll('\r\n', '\n') + .replaceAll('\r', '\n') + .replaceAll('\n', '<br>') + .replaceAll('|', '\\|'); lines.push(`| \`${change.file}\` | ${escapedDesc} |`); }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/pr-ai-description-lib.mjs around lines 556 - 558, Normalize newlines in change.description before escaping pipes: in the loop that builds the table (the variable escapedDesc and the lines.push call), replace CR/LF/R characters (e.g., replaceAll(/\r\n|\r|\n/g, ' ')) on change.description first, then perform the replaceAll('|', '\\|') and any other escaping before pushing the row; this ensures embedded newlines won't split the Markdown table rows.
401-416:⚠️ Potential issue | 🔴 Critical민감정보 마스킹 규칙이 여전히 너무 좁습니다.
Line 401-416의 현재 패턴은
OPENAI_API_KEY,AWS_SECRET_ACCESS_KEY,DATABASE_URL(credential 포함 URL) 같은 흔한 형태를 놓칠 수 있습니다. 외부 전송 전 마스킹 누락 위험이 큽니다.보강 예시
export function maskSensitiveContent(text) { if (typeof text !== 'string' || text.length === 0) { return ''; } return text .replaceAll(/(Authorization\s*[:=]\s*)([^\n\r]+)/gi, '$1[REDACTED]') .replaceAll(/Bearer\s+[A-Za-z0-9\-._~+/]+=*/gi, 'Bearer [REDACTED]') + .replaceAll( + /\b([A-Z0-9_]*(?:API_KEY|ACCESS_KEY|SECRET|TOKEN|PASSWORD|PRIVATE_KEY|CLIENT_SECRET))\b\s*[:=]\s*([^\s,\n\r]+)/g, + '$1=[REDACTED]', + ) + .replaceAll(/(https?:\/\/)([^\/\s:@]+):([^@\/\s]+)@/gi, '$1[REDACTED]:[REDACTED]@') .replaceAll( /(\"?(?:password|secret|token|apiKey)\"?\s*[:=]\s*)\"([^\"\n\r]*)\"/gi, '$1"[REDACTED]"', ) .replaceAll( /(\"?(?:password|secret|token|apiKey)\"?\s*[:=]\s*)([^\s,\n\r]+)/gi, '$1[REDACTED]', ); }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/pr-ai-description-lib.mjs around lines 401 - 416, The current maskSensitiveContent function uses patterns that miss common env var names and credential-containing URLs; update its replacement rules to also match typical secret key names (e.g., OPENAI_API_KEY, AWS_SECRET_ACCESS_KEY, .*_KEY, .*_SECRET, .*_TOKEN, api_key, api-key, client_secret), credential-bearing URLs (e.g., scheme://user:password@host or DB connection strings containing user:pass), and query/string parameters (e.g., ?password=, &api_key=); add or extend replaceAll calls in maskSensitiveContent to include regexes that catch these variants (env var keys, key=value in query strings, and user:pass in URLs) and consistently replace the secret portion with [REDACTED] or "[REDACTED]" while preserving surrounding labels so existing callers still get the masked text.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/scripts/pr-ai-description-lib.mjs:
- Line 253: The reported issue is Prettier formatting mismatches causing ESLint
failures; run the project's Prettier formatter (e.g., prettier --write) over the
code that contains the ternary using typeof file.previous_filename === 'string'
? file.previous_filename : undefined and the other listed similar spots (around
the occurrences at the other noted line ranges) so the ternaries/expressions and
surrounding commas/line breaks match the repo's Prettier config; ensure the
exact expressions (e.g., file.previous_filename usage) are reformatted rather
than manually reformatting to avoid style drift and re-run lint to confirm the
ESLint errors are resolved.
- Line 78: The normalizePath function currently only replaces double-backslash
sequences, leaving single backslashes in Windows paths which breaks glob
matching; update the replacement to target any single backslash globally (e.g.,
use a global regex like /\\/g or replaceAll('\\','/') on filePath) so
filePath.replaceAll/replace converts all '\' to '/' before stripping a leading
'./', ensuring consistent normalized paths across platforms.
---
Duplicate comments:
In @.github/scripts/pr-ai-description-lib.mjs:
- Around line 6-14: The DEFAULT_EXCLUDE_GLOBS array currently omits common
sensitive files so secrets like .env, key/cert files, and credential stores may
be included in diffs; update the DEFAULT_EXCLUDE_GLOBS constant to add patterns
that exclude environment and credential files (e.g., .env, .env.*, *.key, *.pem,
*.p12, *.jks, *.crt, *.der, *.secret, credentials files and common credential
paths like .aws/credentials and .git-credentials) so these sensitive patterns
are never exported by the PR AI description script.
- Around line 556-558: Normalize newlines in change.description before escaping
pipes: in the loop that builds the table (the variable escapedDesc and the
lines.push call), replace CR/LF/R characters (e.g., replaceAll(/\r\n|\r|\n/g, '
')) on change.description first, then perform the replaceAll('|', '\\|') and any
other escaping before pushing the row; this ensures embedded newlines won't
split the Markdown table rows.
- Around line 401-416: The current maskSensitiveContent function uses patterns
that miss common env var names and credential-containing URLs; update its
replacement rules to also match typical secret key names (e.g., OPENAI_API_KEY,
AWS_SECRET_ACCESS_KEY, .*_KEY, .*_SECRET, .*_TOKEN, api_key, api-key,
client_secret), credential-bearing URLs (e.g., scheme://user:password@host or DB
connection strings containing user:pass), and query/string parameters (e.g.,
?password=, &api_key=); add or extend replaceAll calls in maskSensitiveContent
to include regexes that catch these variants (env var keys, key=value in query
strings, and user:pass in URLs) and consistently replace the secret portion with
[REDACTED] or "[REDACTED]" while preserving surrounding labels so existing
callers still get the masked text.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 45315a8b-b4aa-4eac-a6f8-f9897455a64a
📒 Files selected for processing (1)
.github/scripts/pr-ai-description-lib.mjs
| }; | ||
|
|
||
| function normalizePath(filePath) { | ||
| return filePath.replaceAll('\\\\', '/').replace(/^\.\//, ''); |
There was a problem hiding this comment.
normalizePath의 역슬래시 치환이 불완전합니다.
Line 78은 \\(두 글자)만 치환해서 Windows 경로(\)가 그대로 남을 수 있습니다. 로컬 dry-run/Windows 환경에서 glob 매칭이 어긋날 수 있습니다.
수정 예시
function normalizePath(filePath) {
- return filePath.replaceAll('\\\\', '/').replace(/^\.\//, '');
+ return filePath.replaceAll('\\', '/').replace(/^\.\//, '');
}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In @.github/scripts/pr-ai-description-lib.mjs at line 78, The normalizePath
function currently only replaces double-backslash sequences, leaving single
backslashes in Windows paths which breaks glob matching; update the replacement
to target any single backslash globally (e.g., use a global regex like /\\/g or
replaceAll('\\','/') on filePath) so filePath.replaceAll/replace converts all
'\' to '/' before stripping a leading './', ensuring consistent normalized paths
across platforms.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR Summary
GitHub Actions 워크플로우와 PR AI 요약 생성 관련 스크립트/라이브러리/테스트를 추가합니다.
PR 본문에 AI가 생성한 요약 블록을 삽입하고 라벨/담당자/제목(optional)을 자동으로 적용하는 기능을 제공합니다.
로컬 dry-run 스크립트와 유닛 테스트를 포함하여 안전한 실행과 오류 변환/로깅을 처리하도록 구현했습니다.
.github/workflows/pr-ai-description.yml로 PR 이벤트 트리거 워크플로우 추가.github/scripts/pr-ai-description.mjs및 라이브러리pr-ai-description-lib.mjs추가pr-ai-dry-run.mjs와 유닛 테스트 추가로 로컬 검증 및 CI 테스트 제공.coderabbit.yaml)으로 파일별 리뷰 지침 추가package.json에langsmith와openai의존성 추가Changes
.coderabbit.yaml.github/scripts/__tests__/pr-ai-description-lib.spec.mjs.github/scripts/__tests__/pr-ai-description.spec.mjs.github/scripts/pr-ai-description-lib.mjs.github/scripts/pr-ai-description.mjs.github/scripts/pr-ai-dry-run.mjs.github/workflows/pr-ai-description.ymlpr-ai-description.mjs를 실행하도록 구성되어 있습니다.package.jsonlangsmith와openai의존성을 추가하여 OpenAI 호출 및 Langsmith 트레이싱 래퍼를 사용하도록 변경했습니다.Impact
title이 기존 제목과 다르면 PR 제목을 업데이트할 수 있습니다. (라벨ai-title-lock이 있으면 변경되지 않음)chanwoo7)가 추가될 수 있습니다. 퍼미션이 없으면 건너뜁니다.Checklist
GITHUB_TOKEN/OPENAI_API_KEY/LANGSMITH_API_KEY등 시크릿이 올바르게 설정되어 있는지 확인하세요.github/workflows/pr-ai-description.yml의 env 섹션에 자리표시자([REDACTED] secrets.GITHUB_TOKEN }}등)가 남아있는지 확인하고 올바르게 수정하세요.github/scripts/__tests__/*.spec.mjs)가 CI에서 통과하는지 확인하세요 (node --test 실행)pr-ai-dry-run.mjs를 실행해 실제 PR 업데이트 없이 출력된 AI 요약과 블록 형식을 검증하세요maskSensitiveContent동작 포함) 확인하세요 (로그/스텝 서머리 확인)package.json의존성 추가(langsmith,openai) 후 로컬/CI에서 설치 및 빌드가 정상 동작하는지 확인하세요Dependencies
Summary by CodeRabbit
New Features
Tests
Chores