본문으로 건너뛰기
cccskills
무료GitHub에 공개

xhs-automation-publisher

Run authorized Xiaohongshu browser workflows for login checks, creator preview fill, publish blockers, final posting, search, export, or approved engagement. Prefer native signed-in Chrome; keep Python/CDP for deterministic batch fallback.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(15)

  • SKILL.md4.3 KB
  • .gitignore134 B
  • agents/openai.yaml428 B
  • config/accounts.json.example414 B
  • images/publish_temp/.gitkeep0 B
  • LICENSE1.0 KB
  • README.md670 B
  • requirements.txt34 B
  • scripts/account_manager.py9.4 KB
  • scripts/cdp_publish.py188.1 KB
  • scripts/chrome_launcher.py11.0 KB
  • scripts/feed_explorer.py28.4 KB
  • scripts/image_downloader.py6.8 KB
  • scripts/publish_pipeline.py21.5 KB
  • scripts/run_lock.py3.3 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

XHS Automation Publisher

This skill vendors white0dew/XiaohongshuSkills as a deterministic Xiaohongshu batch/export fallback. Interactive logged-in work should first load $browser-automation-router and use the Codex Chrome extension when available. The vendored Python + Chrome DevTools Protocol (CDP) commands remain useful for repeatable preview filling, creator data export, and explicit batch contracts.

Use With The Local Role System

  • Use $xhs-publish-assistant first when the user only needs copy-ready title/body/tags/image directory output.
  • Use $xhs-automation-publisher when the user explicitly needs browser automation, login diagnosis, creator-center filling, auto-publish blocker debugging, content-data export, search/detail collection, or authorized final posting.
  • Keep 内容主编 as coordinator for content strategy and publish authorization. 小红书 executes this skill only after scope and account boundary are clear.

Hard Gates

  • Default to preview/fill, not final posting. Prefer the native Chrome surface for an interactive reviewable fill; use publish_pipeline.py --preview or cdp_publish.py fill when a deterministic script fallback is required.
  • publish_pipeline.py defaults to clicking publish when --preview is absent. Treat that as high risk.
  • Do not run click-publish, publish, post-comment-to-feed, respond-comment, note-upvote, note-unvote, note-bookmark, note-unbookmark, remove-account, re-login, or switch-account without a fresh explicit user approval for that exact action.
  • Do not store cookies, account tokens, login QR payloads, real account names, or Chrome profile paths in this repo.
  • Do not use this skill for scraping or interaction at scale. Respect platform rules, rate limits, and account risk.
  • Stop before final publish if the target account, media list, title, body, tags, or publish time is uncertain.

Surface Selection

  1. Use the Codex Chrome extension for an existing signed-in profile, visible creator-center review, one-off form filling, and user-observed troubleshooting.
  2. Use the in-app Browser only when its separate login state is acceptable; it does not inherit the user's Chrome profile.
  3. Use the vendored Python/CDP commands for batch export, a stable command contract, unattended repeatability, or an explicit fallback when native plugins are unavailable.
  4. Do not launch or copy a local Chrome profile merely to bypass a missing native plugin. Do not commit profile paths or login state.

Safe Default Flow

  1. Confirm target account, desired mode, media files, and whether final posting is authorized.
  2. For interactive work, verify the native Chrome surface and active account. For a script fallback, check/login only when requested:
python scripts/cdp_publish.py check-login
python scripts/cdp_publish.py login
  1. Fill creator-center form for human review:
python scripts/publish_pipeline.py --preview --title-file title.txt --content-file content.txt --images image1.png
  1. Only after explicit approval, run a publish command without --preview or run click-publish.
  2. Return exact commands, status markers, account boundary, and any manual actions left.

Useful Commands

python scripts/chrome_launcher.py
python scripts/chrome_launcher.py --restart
python scripts/cdp_publish.py list-accounts
python scripts/cdp_publish.py search-feeds --keyword "选题关键词"
python scripts/cdp_publish.py content-data --csv-file "content_data.csv"

The scripts require Python 3.10+, Google Chrome, and dependencies from requirements.txt.

Failure Handling

  • Not logged in: switch to headed login and ask the user to scan QR code.
  • Selector/page failure: inspect scripts/cdp_publish.py SELECTORS, upload waits, editor fill, and publish-button logic.
  • Media path failure: use absolute paths; for WSL/remote CDP path mismatch, consider --skip-file-check.
  • Comment/engagement uncertainty: do not guess target IDs or tokens; ask for the exact note/comment target.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Route CEO-first multi-window work into owner-led role loops. Use for 总控/CEO、架构/CTO、内容主编、角色派发与复用、模型路由、来源窗口回调、role-windows 台账、skill 命中统计或 multi-window loop engineering。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

Read-only first workflow for diagnosing application incidents by collecting service, log, network, configuration, and functional evidence before proposing any fix.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

Run user-authorized black-box web security tests and evidence-based reports. Use for 黑盒安全测试、渗透测试、数据泄露、未授权访问、公开 JS/API、登录爆破防护、CORS 或安全响应头检查。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

Choose the narrowest reliable browser surface across Codex Browser, signed-in Chrome, Playwright CLI, and platform scripts. Use for browser automation, logged-in workflows, UI inspection, screenshots, form filling, publish previews, or frontend QA.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

提议并执行 rubric 或 bucket 升级。两种模式:**完整 rubric bump**(最高风险动作,5 步强制 + 跨模型审核)和 **--bucket-only 轻量重校**(只换 bucket 边界,不动 rubric 公式)。**Phase 2 强制走 cheat-score-blind sub-agent 给校准池重打分**——不接受 self-scored fallback。触发词:"升级 rubric"/"bump rubric"/"更新公式"/"我想加一个维度"/"调整权重"/"重校桶"/"recalibrate bucket"。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

初始化 cheat-on-content 内容校准工作区。仅在用户明确开始该工作流,或其子 skill 确认缺少 .cheat-state.json 时使用;不响应一般项目初始化、安装诊断或 Codex 首次会话。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Dirtytrii/codex-skills112026년 10월 9일 업데이트

Dirtytrii의 스킬 모두 보기

이 스킬의 문제 신고