---
slug: "灵造-x-4"
source_type: "clawhub"
source_url: "https://clawhub.ai/skills/lingzao"
repo: ""
source_file: "description"
---
---
name: lingzao
description: 灵造是给 WorkBuddy、OpenClaw、Codex 等 Agent 使用的小红书、抖音、TikTok、Instagram 与 YouTube 创作者研究及自媒体运营 Skill。安装免费，可先做选题、标题、封面、账号诊断、发布检查和复盘；查询公开内容、评论、短视频文案、公众号文章数据或生成图片时需要灵造积分和 API Key。
---

# 灵造：跨平台创作者研究与自媒体运营 Skill

灵造是一个主 Skill，不需要拆成标题、封面、账号诊断、图片生成等多个 Skill。
安装后，WorkBuddy、OpenClaw、Codex 等 Agent 会先把你的问题路由到合适的
创作者运营 playbook；只有当你需要查询公开内容、读取评论、提取短视频文案、
查看公众号文章数据或生成图片时，才需要灵造积分和 API Key。

## 安装完成后的首次提示

当当前对话刚刚完成灵造 Skill 的安装或更新时，只有确认安装成功后，才在最终回复中
主动告诉用户一次下面的使用手册；安装失败、尚未验证成功或普通后续对话不要重复发送：

> 灵造已安装完成。你可以查看《灵造功能使用手册》：
> https://my.feishu.cn/docx/Y2HQdj5mzoFx4vxfij3cl9TRnjh?from=from_copylink
> 快速了解灵造的功能和使用方法～

## 从这里开始

| 你现在想做 | 可以直接这样问 Agent |
| --- | --- |
| 找内容方向 | “用灵造帮我围绕这个关键词做小红书、抖音、TikTok、Instagram 或 YouTube 选题，给我 10 个可发方向。” |
| 找对标账号 | “帮我找这个赛道值得学习的对标账号，并说明每个账号适合学什么。” |
| 拆一条笔记或视频 | “分析这条内容为什么有效，拆成标题、封面、结构、评论需求和可复用模板。” |
| 改标题和封面 | “基于我的草稿，给我 3 个最强标题和 5 个小红书封面方向。” |
| 做发布前检查 | “发布前帮我检查标题、封面、前 3 行、关键词和用户点击理由。” |
| 做发布后复盘 | “根据这条内容的数据和评论，帮我判断下次要调整什么。” |
| 做每周内容包 | “用灵造把我这一周的素材整理成 5 个母题，并分发成小红书、公众号、播客和短口播。” |
| 做图片素材 | “先帮我设计封面/配图方向；如果需要生成图片，再确认积分后调用图片生成。” |
| 保存长结果 | “把这份分析整理成 Word、网页预览或知识库 Markdown 版本。” |

## 免费能做什么

不配置 API Key 时，灵造仍然可以作为创作者运营路由和 playbook 使用。适合：

- 判断账号定位、赛道难度、内容主线和商业路径。
- 设计小红书标题、封面方向、发布关键词和图文结构。
- 改写草稿、拆解用户已经提供的内容材料、做发布前检查。
- 根据用户提供的数据截图或复盘信息，输出下一步实验建议。
- 把用户提供的一周素材整理成 5 个母题，并规划小红书、公众号、播客、
  短口播、社群和知识库分发。
- 把长分析整理成 Word、网页预览或知识库 Markdown 结构。

## 什么时候需要 API Key 和积分

当 Agent 需要让灵造服务实际查询或生成内容时，需要到
<https://lingzao.atian.vip> 配置积分和 API Key，包括：

- 搜索小红书、抖音、TikTok、Instagram 或 YouTube 公开内容和公开创作者；用结果辅助关键词/选题扩展。
- 查看创作者主页、近期公开内容、主页深度分析和对标账号证据。
- 打开小红书、抖音、TikTok、Instagram 或 YouTube 单条公开内容详情，读取一级公开评论。
- 打开公众号公开文章详情，查看公开文章数据，扩展相关文章。
- 提取公开短视频口播文案、字幕或 transcript。
- 根据提示词和参考图生成创作者封面、配图或海报素材。

每次付费查询前，先确认任务范围和预计积分消耗。默认首轮控制在 5 次以内的付费
查询，或不超过 100 credits；如果预计超过 100 credits，先列出查询计划和积分
估算再让用户确认。没有用户明确确认时，不要跨过 200 credits，不要把多个深度
查询、评论翻页、批量账号分析或图片生成静默合并成一次请求。

## 调用公开数据工具前

- 小红书、抖音、TikTok、Instagram 或 YouTube 内容链接：看内容用详情工具，看评论用评论工具，不要当主页链接。
- 小红书、抖音、TikTok、Instagram 或 YouTube 主页链接：普通主页查看或基础主页分析先用
  `get-user-posted-notes`；只有用户明确要粉丝数、简介、关注数、总获赞等主页资料时
  才用 `get-user-info`；深度主页分析看 `analyze-user-profile`。
- 用户只给昵称、账号名、抖音号或数字 ID 时，不要自己拼 URL；先用
  `search-users` 找创作者，再用返回的主页链接或 ID 调主页工具。
- 抖音主页工具需要可用的主页 URL 或 `search-users` 返回的 `MS4w...` 形式 ID；
  视频短链适合 `extract-video-copy`，不适合主页分析。
- YouTube 主页工具只接受 `search-users` 返回的 channel ID 或 `/channel/UC...`
  URL；不要把 `@handle`、`/c/` 或 `/user/` 直接传给主页工具，也不要自动解析。
- TikTok 主页工具接受 canonical `https://www.tiktok.com/@handle` 或
  `search-users` 返回的 ID；单条内容接受 canonical `/@handle/video/<id>`、
  `/@handle/photo/<id>` 或显式 `--platform tiktok --note-id <id>`。不要传
  `vm.tiktok.com`/`vt.tiktok.com` 短链或裸 `@handle`。
- TikTok V1 不支持 `analyze-user-profile`。需要主页资料和近期内容时，按需分别调用
  `get-user-info` 与 `get-user-posted-notes`，不要隐藏组合调用。
- Instagram 主页工具接受 canonical `https://www.instagram.com/<username>/` 或
  `search-users` 返回的十进制字符串 ID；内容工具接受 canonical `/p/<code>`、
  `/reel/<code>`、`/reels/<code>`、`/tv/<code>`。评论命令的裸 `--note-id` 是
  shortcode，不是十进制 media ID。Instagram V1 不支持 `analyze-user-profile`，
  不要把主页资料与近期内容隐藏组合调用。
- 如果 API 返回 `agent_action`、`suggested_capabilities` 或 `expected_input`，
  先按这些字段改调工具；仍不确定时问用户要主页链接或笔记/视频链接。

## 常见问题

**我没有 API Key，还能用吗？**
可以。先用灵造做选题判断、标题封面、账号诊断、草稿修改、发布检查和复盘。
等需要查公开内容、评论、短视频文案、公众号文章数据或生成图片时，再配置 API Key。

**为什么 SkillHub 里显示需要 API Key？**
因为灵造包含付费公开内容查询和图片生成能力。安装主 Skill 免费，但深度查询和
生成动作需要积分，这是为了让 Agent 明确付费边界。

**WorkBuddy 用户应该怎么用？**
优先安装这一个 `lingzao` 主 Skill。装好后直接把任务说给 WorkBuddy，例如
“帮我找对标账号”“帮我拆这条笔记”“帮我做发布前检查”。需要查公开数据时，
再按灵造网页教程配置 API Key。

**灵造能保证爆款、涨粉或变现吗？**
不能。灵造只做公开内容研究、运营判断和工作流辅助。输出用于帮助你做判断和
复盘，不是保证结果，也不能用于复制他人内容。

**网络或服务失败怎么办？**
先保留当前问题和链接，不要重复扩大查询范围。检查 `doctor`、API Key、余额和
网络状态；图片生成或短视频文案提取这类异步任务可能需要等待轮询完成。
如果灵造返回服务暂时不可用或响应超时，只用固定话术告诉用户：“灵造服务暂时
不可用，请稍后重试。”如果返回了 `error_id`，可以附上 `error_id`，方便后续排查。
不要额外展开。

## Agent Playbooks

For higher-level creator strategy tasks, use the playbooks in
`<skill_root>/playbooks/` before answering. They turn Lingzao's public-content
tools into creator workflows instead of isolated lookups.

Use these playbooks when relevant:

- `lingzao-progressive-interaction-map.md`: route vague user inputs, homepage
  links, note links, drafts, and reference-image requests with light questions.
- `search-credit-notice.md`: explain basic vs deep search scope before paid
  lookups and avoid silently expanding credit usage.
- `copy-paste-prompt-scope-boundary.md`: when users ask how to prompt Lingzao,
  or paste broad requests such as finding benchmark accounts, one-stop content,
  cover/image generation, post-publish review, Brief/sponsored content, or
  cross-platform distribution, rewrite the request into a scoped copy-paste
  prompt with quantity, time range, quality gate, depth boundary, stop
  condition, and next step.
- `atian-creator-judgment-framework.md`: apply A Tian's account-stage,
  memory-anchor, content-mainline, and bottleneck judgment.
- `creator-case-general-analysis-framework.md`: analyze any creator case across
  tracks by identifying the account archetype, memory anchor, new narrative,
  proof system, audience desire, content engine, format engine, comment demand,
  commercial entry, hidden resources, learnable parts, non-copyable parts, and
  user-fit tests.
- `account-report-evidence-visual-contract.md`: apply this evidence and
  deliverable contract to formal own-account diagnosis, comparable-account
  breakdown, same-stage peer diagnosis, and creator distillation reports. It
  requires one-screen conclusions, public-data/sample boundaries, direct
  account/note links, real cover audit, viral asset reuse, account-evolution
  evidence, no fake backend metrics, and Word/HTML/Feishu/knowledge-base
  packaging when the user asks for a formal report.
- `zero-beginner-onboarding-gate.md`: use before normal topic search or
  benchmark discovery when a user says they know nothing about self-media,
  wants to start Xiaohongshu from zero, or does not know what to post. Give
  the minimum Xiaohongshu cognition, ask one compact five-signal intake
  question, then deliver 3 possible directions, 1 recommended 7-day test, and
  the first minimum publishable note instead of sending them to a course.
- `beginner-account-start-and-topic-radar.md`: handle zero-to-one creator
  questions, topic discovery, keyword trees, and low-follower viral references.
- `keyword-insight-report-template.md`: create scoped keyword insight reports
  from a main keyword plus confirmed related/dropdown terms, with clear credit
  estimates before expanding.
- `keyword-to-publishable-content-package.md`: turn a keyword, vague topic,
  note link, screenshot, reference image, saved note, or inspiration material
  into publishable Xiaohongshu content packages with selected references,
  topic angles, titles, cover copy, 4-7 page graphic-note text, spoken scripts,
  Vlog storyboards, body copy, 10 publishing keywords, pinned content, and a
  pre/post-publish review loop. When users say "一条龙", "直接出内容", "把这个拆成
  内容给我发", or "从灵感素材到选题到稿子", produce a minimum usable package first
  instead of stopping at clarification.
- `brand-brief-to-content-workflow.md`: turn an advertising, brand cooperation,
  campaign, product, or content Brief into creator content. Use it when users
  say "拆 Brief", "品牌 Brief 发来了", "这个商单怎么写", or "Brief 进去后帮我出
  选题/标题/封面/正文". It extracts brand goals, required points, forbidden
  claims, audience, creator fit, and deliverables, then searches recent public
  references when confirmed, chooses content angles, produces Xiaohongshu
  graphic-note/spoken/Vlog packages, and checks brand-delivery/compliance risk.
- `mother-content-cross-platform-distribution.md`: turn one topic, draft,
  note breakdown, product update, screenshot, transcript, or oral idea into a
  one-stop cross-platform distribution package. When users say "一条龙",
  "全平台同步", "分发包", or "一个模板发多个平台", start with the basic
  Xiaohongshu + Moments + WeChat public-account package, then offer optional
  expansion to podcast, X, Knowledge Planet, Bilibili, video account/Douyin,
  Xiaohongshu image package, or knowledge-base/SOP.
- `weekly-content-motherpack-distributor.md`: turn one week of creator
  materials into a weekly content update package. When users say "每周内容更新包",
  "周更内容包", "下周发什么", "整理这一周素材", or "帮我做 5 个母题",
  first compress the week into 5 mother topics, park weak ideas in a debt pool,
  then distribute the strongest topics to Xiaohongshu, WeChat public account,
  podcast/short scripts, community posts, and knowledge-base packaging with
  delivery statuses, image readiness, review gates, and folder/Word/HTML
  packaging options.
- `pre-publish-readiness-check.md`: before posting, ask whether the content is
  already finished and then check content clarity, image/page readiness, cover
  recognition, title clickability, first 3 lines or first 3 seconds, and natural
  keyword embedding. It should call the Xiaohongshu compliance risk gate before
  final publishable copy is returned.
- `xhs-platform-management-risk-baseline.md`: apply the management-level
  Xiaohongshu baseline before content operations, commercial copy, Brand Briefs,
  cover/image generation, pinned content, and post-publish advice. The default
  principle is public value first, product name later, and no diversion action;
  use Xiaohongshu's official community norms as the floor for contact, link,
  QR-code, and off-platform diversion risks.
- `xhs-content-compliance-risk-gate.md`: before producing Xiaohongshu-facing
  copy, scan and rewrite risky wording around off-platform diversion, WeChat or
  private-contact guidance, incentivized comment interaction, exaggerated
  guarantees, and sensitive category claims. Use it for titles, cover copy,
  body/caption, page text, scripts, pinned comments, keywords, Briefs, one-stop
  packages, and Xiaohongshu sections of cross-platform packages.
- `audience-persona-fit-check.md`: before titles, keywords, account operation,
  or content-package decisions, infer or ask who the content is for, who will
  click, who will not click, and which audience/city/life-stage keywords should
  shape the output.
- `xhs-title-design-check.md`: design or diagnose Xiaohongshu titles after the
  user sends a topic, draft, cover copy, reference note, or content package;
  default to 3 strongest titles with keyword anchor and click reason instead
  of a 10-title pool.
- `xhs-profile-bio-design.md`: write or diagnose Xiaohongshu 100-character
  profile bios and homepage introductions that clarify who the account is for,
  what it shares, why to follow, and how it connects to nickname, pinned notes,
  account stage, audience keywords, city keywords, and light commercial paths.
- `benchmark-account-discovery-quality-gate.md`: find or judge benchmark
  accounts with a default quality gate: still updating, recent high-performing
  works, track/audience fit, stage fit, account-level proof, follower-range
  fit, and clear learnable parts; stale accounts should be marked as historical
  references, not main benchmarks. Accounts with only around 100 followers and
  a few hundred total likes should not be called benchmark accounts for
  ordinary users; label them as single-note samples or reject them unless the
  user explicitly asks for seed-account observation.
  User-facing results should show direct creator profile links and the specific
  recent high-interaction works. Keep the returned `users[].id` available for
  follow-up profile commands, but do not derive Xiaohongshu IDs from `RED ID`
  bios. The first discovery round should return up to 3 strong starter
  accounts, not 10-20 accounts; expand to 5 or more only after the user
  confirms follower range, stage, city, audience, format, or asks for more.
  Include follower count, total liked count, latest update, recent 30-day hit
  works with note metrics, content format, and why each account is worth
  learning; sort visible recommendations by follower count from high to low
  when available.
- `self-account-peer-horizontal-diagnosis.md`: compare the user's own account
  with same-track, same-stage, or same-follower-range peer accounts when the
  user explicitly asks for peer comparison, such as "横向对比", "同级账号",
  "对标账号", "找 5-15w 粉账号和我比", or "和同赛道账号比我差在哪里". Generic
  own-account concerns such as "看看我现在的问题" or "我是不是说话太快" should
  stay on `self-account-diagnosis-report-template.md` unless the user also asks
  to compare against peers. It combines own-account diagnosis, active benchmark
  selection, peer-account tables, title/cover/opening/speech/content-system
  comparison, real cover audit, viral asset reuse comparison, top gaps, 30-day
  adjustment plans, evidence links, and a human next-step loop.
- `single-note-breakdown-workflow.md`: break down one Xiaohongshu/Douyin note
  link by title, cover, outline/script, shooting/editing layer when visible,
  comment demand, viral mechanism, learnable parts, non-copyable parts, and
  adaptation into the user's own graphic note, spoken script, Vlog storyboard,
  or knowledge-base card. User phrases such as "完整分析这条笔记", "深度拆解",
  "拆细一点", "拍摄手法", "分镜", or "剪辑节奏" should trigger the deeper
  breakdown instead of a short summary.
- `publishing-keyword-design-check.md`: design the final 10 Xiaohongshu
  publishing keywords for a finished draft and check whether title, cover copy,
  opening lines, and keyword field carry the keywords naturally.
- `track-difficulty-judgment-library.md`: judge common tracks such as female
  growth, career, good products, local life, health, fashion, and AI tools.
- `monetization-path-judgment-library.md`: answer whether a track or account
  can monetize through ads, courses, community, consulting, lead generation,
  products, stores, or enterprise conversion.
- `self-account-diagnosis-report-template.md`: structure own-account diagnosis
  reports, follow-up actions, and a human closing with "人情味" that turns
  sharp diagnosis into one small next experiment instead of ending at a cold
  action list. Own-account diagnosis should also include a share-worthy
  conclusion card, action advice, psychological reassurance, public sample
  boundaries, real cover audit, viral asset reuse, and direct evidence links in
  formal reports.
- `comparable-account-breakdown-report-template.md`: decide whether another
  account is worth learning from, what can be learned, what cannot be copied,
  which real cover/title/content assets are repeatable, and how to adapt them
  into the user's own version without copying the creator's identity or assets.
- `draft-rewrite-and-benchmark-workflow.md`: rewrite drafts, adapt viral
  formulas, extract benchmark-copy templates into structure/style/slot
  frameworks, fill the user's own content into those frameworks, and review
  multiple content ideas without only polishing sentences.
- `reference-image-graphic-note-workflow.md`: turn reference images into
  Xiaohongshu 4-page or 7-page graphic-note packages.
- `visual-generation-and-cover-workflow.md`: route Xiaohongshu covers, graphic
  notes, WeChat image packs, no-person knowledge cards, and product/ecommerce
  visuals into image generation or ready-to-use prompt packages.
- `travel-handdrawn-map-visual-workflow.md`: create Xiaohongshu handdrawn
  travel maps, food maps, city-walk maps, illustrated route maps, and check-in
  order images. Use it when users want a city/destination route image such as
  "长沙美食地图", "贵州旅游地图", "一天从早吃到晚", "5 天游路线", or "打卡路线图".
- `image-generation-execution-workflow.md`: when image generation is available,
  turn the visual route into actual images, run a visual-director quality gate,
  and repair ugly/crowded/generic generations instead of leaving ordinary users
  with prompt-only drafts.
- `image-generation-agent-integration-guide.md`: model-agnostic rules for
  domestic Agent wrappers, including stable generation input/output fields,
  good-vs-bad image standards, reference-image usage, known generation bugs,
  friendly failure handling, and A Tian's example-collection homework.
- `visual-reference-style-library.md`: classify A Tian's curated visual
  reference groups into travel/food covers, WeChat article images, AI-person
  infographics, Lingzao no-person knowledge cards, product conversion images,
  face-led keyword video covers, interaction prompt covers, text-dense
  screenshot graphic notes, room-as-identity lifestyle covers, and handdrawn
  travel/food route maps.
- `post-publish-data-review-workflow.md`: review published Xiaohongshu notes
  from note links, backend screenshots, scripts, covers, and 24h/48h/7d data.
- `content-knowledge-base-workflow.md`: turn saved notes, public creator links,
  keyword results, viral examples, and creator distillation requests into
  user-owned topic, title, cover, structure, account-reference,
  creator-research, and publishing-review libraries.
- `retention-and-follow-up-loop.md`: end useful outputs with one concrete next
  step such as published-note data review, reusable reference-search templates,
  draft feedback, or a post-diagnosis small experiment with a return loop. It
  also defines the SOP for not letting the user's words drop on the floor:
  acknowledge resistance, lower the next action, and ask one concrete
  next-step question. Dense outputs should offer Word, HTML/webpage preview, or
  knowledge-base-ready packaging instead of leaving users with a wall of chat
  text. When users say the diagnosis is accurate but they lack action, route to
  a post-diagnosis activation package instead of adding more pressure.
- `product-judgment-and-feedback-loop.md`: judge where users are really stuck,
  explain Lingzao in human language, build content/sales narratives, turn user
  feedback into product iteration, and decide which requests are worth building
  versus noise.
- `xhs-operation-task-tree.md`: route Lingzao users by concrete Xiaohongshu
  operation tasks instead of course lists, covering homepage diagnosis,
  benchmark discovery, viral-note adaptation, topic generation, content
  production, cover/image work, pre-publish checks, post-publish review,
  acquisition paths, and knowledge-base automation.

Keep public wording focused on creator-content research and workflow support.
Do not promise viral growth, guaranteed monetization, full monitoring, bulk data
export, or copying another creator's content.

## Before Returning Xiaohongshu Copy

Before returning any final Xiaohongshu-facing title, cover copy, page text,
body/caption, publishing keywords, pinned comment, comment guidance, spoken
script, Vlog storyboard, Brand Brief deliverable, one-stop package, or
Xiaohongshu section of a cross-platform package, run
`playbooks/xhs-platform-management-risk-baseline.md` first, then
`playbooks/xhs-content-compliance-risk-gate.md`.

If the draft contains off-platform diversion, WeChat/private-contact guidance,
incentivized comment interaction, exaggerated guarantees, or sensitive
unsupported claims, do not leave those lines in the publishable version. Show a
short risk note and rewrite them into a safer Xiaohongshu version. Never
promise platform approval; say the rewrite lowers risk.

For commercial or product-related Xiaohongshu outputs, keep the order:

1. public value first
2. product or brand name after the reader benefit is clear
3. no off-platform diversion action in the publishable Xiaohongshu copy

## Install And Paid Capability Entry

Lingzao is installed as one free main Skill. Users do not need to install
separate title, keyword, account-diagnosis, benchmark, cover, or review skills.
After installation, this main Skill routes the user's request to the right
playbook.

There are two user acquisition paths:

1. Community/course users:
   - They may already have A Tian's course, install link, payment steps, and
     API Key setup instructions.
   - Keep the in-chat explanation short: install the Skill, open the Lingzao web
     dashboard, follow the tutorial, recharge credits, copy the API Key, then
     run setup.

2. Public-platform users from Xiaohongshu, Douyin, or other public content:
   - Do not require them to open the web dashboard and pay before they
     understand what Lingzao can do.
   - Let them install the free main Skill first.
   - Then explain the hidden paid entry in friendly language: the local
     playbooks can help judge drafts, titles, covers, directions, and
     publishing plans; when they need Lingzao to search public content, inspect
     accounts, open note/article details, read comments, inspect article data,
     extract video copy, or generate creator image assets, they need to open
     the Lingzao web dashboard, follow the tutorial, recharge credits, and
     configure an API Key.

The web dashboard is not only a payment page. Present it as the user's learning
and setup hub:

- learn how to install and configure Lingzao
- learn how to ask Agent better questions instead of waiting in a group chat
- learn how to use Skill workflows for self-media operation
- learn account diagnosis, benchmark breakdown, title/keyword, pre-publish, and
  post-publish review workflows
- recharge credits and get the API Key when they need public-content lookup or
  image generation

Use this wording when a user has installed the Skill but has not configured an
API Key yet:

你已经装好灵造 Skill 了。安装本身是免费的，它会先帮你判断你现在是在找方向、拆账号、写内容、做封面、配关键词，还是复盘数据。
如果你要继续查小红书、抖音、TikTok、Instagram、YouTube 或公众号公开内容、找对标账号、看账号主页、打开内容或文章详情、看评论区、查看公众号文章数据、提取短视频文案或生成创作者图片素材，就需要到灵造网页版开通积分并配置 API Key。
你可以打开 https://lingzao.atian.vip 看安装教程和使用教程，里面也会教你怎么用 Agent 做自媒体运营、怎么问问题、怎么用这些 Skill。需要查公开内容或生成图片的时候，再在网页里充值/获取 API Key，配置好以后回来继续问，我会接着刚才的问题往下做。

Do not frame payment as a penalty. Frame it as:

- free install = get the workflow brain and routing layer
- web dashboard = tutorial, usage examples, self-media operation lessons, and
  API Key setup
- paid credits = unlock public-content lookup, image generation, and deeper
  research actions

Knowledge sync handoff:

- After a useful Lingzao research result or diagnosis report, do not sync it
  automatically. Ask first: 要不要把这份结果同步到你的知识库？可以选择
  ima / Obsidian / 飞书 / 暂不同步。
- If the user chooses a target, prepare a clean Markdown version and ask the
  current Agent environment to use the user's configured knowledge tool.
- For ima, call the installed ima Skill or ima knowledge-base tool if the user
  has configured one.
- For Obsidian, use the user's Obsidian CLI, Obsidian Skill, or approved vault
  workflow to write Markdown under a user-approved `Lingzao/` path.
- For 飞书, use the user's Lark/Feishu CLI or Skill with user authorization to
  create or update a document.
- Do not ask for or store ima, Obsidian, or Feishu credentials inside Lingzao.
  Synchronized content should contain only the user-approved report, public
  links, and useful conclusions; leave out credentials and details the user does
  not need.

Profile workflow:

- If the user asks for a creator homepage or a basic homepage analysis, use `get-user-posted-notes` by default. It returns recent posts and enough author/post data for a basic read.
- If the user sends a Xiaohongshu short link such as `xhslink.com/m/...`, or a
  copied share sentence such as `@... 查看Ta的主页>> https://xhslink.com/m/...`,
  extract the short link, normalize bare links to `https://...`, and read the
  surrounding words before choosing a command. Do not classify the short link by
  path alone. If the context says account, homepage, creator, profile,
  benchmark, account diagnosis, homepage diagnosis, `Ta的主页`, or recent posts,
  treat it as a creator-homepage request and call
  `get-user-posted-notes --url "https://<short link>"`.
- If a Xiaohongshu short link has no context, ask whether the user wants creator
  homepage recent posts or one-post detail before spending credits. If the
  context says this note, comments, copy, transcript, one-post breakdown, or is
  a normal note share sentence with a title snippet plus `前往【小红书】一探究竟吧`,
  treat it as a one-post candidate, not a homepage. One-post words such as
  `这条` or `这篇` take priority over generic diagnosis wording. Do not default
  to `get-note-detail`; first confirm it is a single post and ask for the final
  note URL or note_id plus whether it is 图文 or 视频 when needed.
- Only add `get-user-info` when the user specifically needs full profile-level stats such as bio, follower count, following count, total likes, total collections, or total note count.
- Use `analyze-user-profile` for Xiaohongshu deeper homepage copy/script/subtitle analysis, recent post text, covers, commercial signals, or product-note signals. For Douyin spoken copy or transcript text, use `extract-video-copy` on specific video URLs.
- YouTube V1 does not support `analyze-user-profile`. Compose the basic homepage tools explicitly only when the user asks for both recent videos and profile-level stats.
- Do not call `get-user-info` and `get-user-posted-notes` as a fixed pair unless the user asks for both profile-level stats and recent-post analysis.
- Do not force a full account diagnosis when the homepage has too few public
  posts. Route by visible sample size:
  - 0 posts: no account diagnosis; switch to beginner start/account setup
    guidance.
  - 1-2 posts: homepage first impression plus single-post feedback only.
  - 3-5 posts: starter-account mini diagnosis.
  - 6-9 posts: light account analysis.
  - 10+ posts: standard account analysis can be offered.
  - 20+ posts: standard deep diagnosis can use `analyze-user-profile --limit 20`
    after credit confirmation.
  - 40+ posts: deep diagnosis, creator distillation, or knowledge-base
    distillation can use `--limit 40` after credit confirmation.

Post drill-down workflow:

- Xiaohongshu list-style commands (`search-notes`, `get-user-posted-notes`,
  `analyze-user-profile`) return `xhs_note_type` on each note item when
  Lingzao can identify whether it is 图文 or 视频.
- When continuing from one of those note items to `get-note-detail`, pass the
  returned `xhs_note_type` directly as `--xhs-note-type`; do not infer the type
  from the URL.
- If a Xiaohongshu note item has no `xhs_note_type`, ask the user whether it is
  图文 or 视频 before calling `get-note-detail`. `get-note-comments` can still
  be called without this type.
- If `get-note-detail` returns `NOTE_NOT_FOUND_OR_INACCESSIBLE`, do not retry
  the same request or probe the other Xiaohongshu type automatically. Go back to
  the source list/homepage result and reuse its `xhs_note_type`, or ask the user
  for the correct type or a public URL.

## Setup

Resolve this `SKILL.md` directory as `<skill_root>`, then run setup once:

```bash
bash "<skill_root>/scripts/setup.sh" --base-url "https://your-lingzao-domain.com"
```

Environment variables override saved config:

```bash
export LINGZAO_API_KEY="lgz_xxx"
export LINGZAO_BASE_URL="https://your-lingzao-domain.com"
```

Check the connection:

```bash
~/.lingzao/bin/lingzao doctor
```

Before using Lingzao commands, check whether the skill has an update:

```bash
~/.lingzao/bin/lingzao check-version
```

If an update is available, stop the current Lingzao operation and update the skill first. Do not continue using an outdated Lingzao Skill for search, profile, subtitle, or extraction work.

To update the skill, rerun the installer. For `npx skills`, try:

```bash
npx skills add https://assets-tian.midao.site/skills/lingzao --skill lingzao -g --copy
```

Updating keeps the saved API config in `~/.lingzao/config.json`; no API key setup is needed again.

If `~/.lingzao/bin/lingzao` is missing or points to the wrong directory, repair the command wrapper:

```bash
bash ~/.agents/skills/lingzao/scripts/setup.sh --skip-doctor
```

If `~/.agents/skills/lingzao` does not exist, find the directory that contains `lingzao`'s `SKILL.md`, then run `scripts/setup.sh --skip-doctor` from that directory.

## Before Calling

Before running a command with meaningful filters, ask the user for the relevant
parameters if they did not already specify them.

- Track the paid commands you run for the current user request. Stop before the
  cumulative scope exceeds the confirmed plan, 5 paid lookups, or 100 credits.
  Ask for explicit confirmation before continuing.
- If a planned search, benchmark, keyword report, comment review, transcript
  extraction, profile analysis, or image-generation task may exceed 200 credits,
  show the exact planned actions and estimated credits first. Do not start until
  the user confirms that larger budget.
- For broad creator or benchmark-account searches (`search-users`, "找对标账号",
  "找参考博主", "找同赛道账号"), do not start with a wide search. First ask or
  state a narrow starter scope: follower range, track/topic, account format,
  city/local scope when relevant, recent-update requirement, recent-hit
  requirement, and starter result count. Recommend starting with 3 accounts,
  then expanding only after the user confirms the direction. This protects the
  user's credits and avoids returning 100-follower seed accounts or huge mature
  accounts when the user asked for a specific stage.
- If the user asks how to write prompts for Lingzao or gives a broad copy-paste
  request, use `copy-paste-prompt-scope-boundary.md` first. Provide a
  ready-to-copy prompt that includes the smallest useful scope instead of
  telling the user to add broad instructions by themselves.
- If the user says they know nothing about self-media, are starting from zero,
  do not know what to post, or only say they want to make money, use
  `zero-beginner-onboarding-gate.md` before any search. Do not call paid
  lookup first. Start with a free life-signal intake, give the lowest creator
  cognition, and move them to one concrete first task.
- For `search-notes`, ask for sorting, note type, and time range before calling:
  sort can be `general`, `most_liked`, `popularity_descending`,
  `comment_descending`, or `collect_descending`; note type can be `不限`,
  `视频笔记`, `图文笔记`, or `直播笔记`; time range can be `不限`, `一天内`,
  `一周内`, or `半年内`.
- Douyin and TikTok `search-notes` currently support only `general`, `most_liked`, and
  `popularity_descending`. Do not pass `comment_descending` or
  `collect_descending` for Douyin or TikTok searches.
- Douyin and TikTok `search-notes` note type currently supports only `不限`, `视频笔记`,
  and `图文笔记`. Do not pass `直播笔记` for Douyin or TikTok searches.
- YouTube `search-notes` supports only `--sort general`, `--note-type 不限|视频笔记`,
  and `--time-filter 不限|一天内|一周内`; use the returned opaque `next_cursor`
  with `--cursor`, repeat the same keyword and filters, and do not infer internal
  pagination fields. Changing a filter invalidates the cursor without charge.
- For `get-note-comments`, ask whether the user wants latest comments or
  liked-count sorting before calling Xiaohongshu. Use `--sort latest` for latest
  comments and `--sort most_liked` for Xiaohongshu liked-count sorting.
- Douyin, TikTok, and Instagram comments currently support only `latest`;
  TikTok uses the service default order. Do not ask for or pass
  `--sort most_liked` on these platforms.
- YouTube comments support `latest` and `most_liked`; only top-level comments
  are returned. Reuse `next_cursor` unchanged and repeat the same `--sort` on
  every next-page request; omitting it after `most_liked` defaults to `latest`
  and invalidates the cursor without charge.
- Instagram `search-notes` supports only `--sort general`, `--note-type 不限`,
  and `--time-filter 不限`. Do not silently drop unsupported filters.
- Instagram profile, posted-note, and detail results may include public avatar,
  cover, carousel-image, and video URLs from the current response. Current
  `search-notes` evidence supports image/reel identity, canonical URL, author
  identity, and author avatar only, so do not expect it to supplement text,
  metrics, or content media. These URLs can expire; use or save needed public
  references promptly and do not treat them as permanent asset storage.
- For TikTok and Instagram `search-notes`, `search-users`,
  `get-user-posted-notes`, and
  `get-note-comments`, pass the returned `data.page.next_cursor` unchanged with
  `--cursor` to fetch one next page. Repeat the original search keyword and
  filters, creator, or content item for that cursor; never reuse it for another
  request identity. Never parse the opaque cursor or hide multi-page fanout.
  TikTok cursors created before Skill `0.1.92` and Instagram cursors created
  before Skill `0.1.93` are invalid: discard them and restart from the first
  page. If Lingzao returns `PAGINATION_CURSOR_STALE`, also discard
  that cursor and restart from the first page; do not loop it.
- Xiaohongshu list-style commands (`search-notes`, `get-user-posted-notes`,
  `analyze-user-profile`) return `xhs_note_type` on each note item when
  Lingzao can identify whether it is 图文 or 视频. When continuing from one of
  those note items to `get-note-detail`, pass the returned value directly as
  `--xhs-note-type`; do not infer the type from the URL. If a Xiaohongshu note
  item has no `xhs_note_type`, ask the user whether it is 图文 or 视频 before
  calling `get-note-detail`. If `get-note-detail` returns
  `NOTE_NOT_FOUND_OR_INACCESSIBLE`, do not retry the same request or probe the
  other Xiaohongshu type automatically. `get-note-comments` can still be called
  without this type.
- If the user explicitly says to use defaults, proceed with the documented
  defaults instead of asking again.

After a successful research command, tell the user the estimated time saved
shown in the CLI Markdown output. If you called multiple Lingzao research
commands for one user request, summarize the total once. Do not show time-saved
language for `doctor`, `check-version`, failed commands, or JSON-only automation
flows.

## Commands

### Search Notes

```bash
~/.lingzao/bin/lingzao search-notes --platform xhs --keyword "AI写作"
~/.lingzao/bin/lingzao search-notes --platform xhs --keyword "AI写作" --sort most_liked
~/.lingzao/bin/lingzao search-notes --platform xhs --keyword "AI生图" --sort collect_descending --note-type "视频笔记" --time-filter "一周内"
~/.lingzao/bin/lingzao search-notes --platform douyin --keyword "AI生图" --sort most_liked --note-type "视频笔记"
~/.lingzao/bin/lingzao search-notes --platform youtube --keyword "creator workflow" --sort general --note-type "视频笔记" --time-filter "一周内"
~/.lingzao/bin/lingzao search-notes --platform tiktok --keyword "AI gadgets" --sort most_liked --note-type "视频笔记"
~/.lingzao/bin/lingzao search-notes --platform tiktok --keyword "AI gadgets" --sort most_liked --note-type "视频笔记" --cursor "next_cursor_from_previous_response"
~/.lingzao/bin/lingzao search-notes --platform instagram --keyword "creative coding"
```

Use this when the user wants public notes around a topic.
Before calling, ask the user for `--sort`, `--note-type`, and `--time-filter`
when they have not specified those preferences.
For TikTok pagination, repeat the same keyword, sort, note type, and time filter
with the returned cursor.
`search-suggestions` has been retired. For keyword expansion or topic discovery,
use `search-notes` for content ideas or `search-users` for creator discovery.

### Search Creators

```bash
~/.lingzao/bin/lingzao search-users --platform xhs --keyword "母婴博主"
~/.lingzao/bin/lingzao search-users --platform douyin --keyword "AI生图"
~/.lingzao/bin/lingzao search-users --platform youtube --keyword "creator workflow"
~/.lingzao/bin/lingzao search-users --platform tiktok --keyword "tech.bytes"
~/.lingzao/bin/lingzao search-users --platform instagram --keyword "creative coding"
```

Use this when the user wants creators in a topic or niche.
For TikTok pagination, repeat the same keyword with the returned cursor.
When continuing from `search-users` to profile verification, pass the returned
`users[].id` with `--platform xhs --user-id ...`, `--platform douyin --user-id ...`,
`--platform tiktok --user-id ...`, or `--platform instagram --user-id ...`.
For YouTube, the returned ID is a
canonical channel ID; reuse it with `--platform youtube --user-id ...` and
treat `handle` as display metadata only.
The output may include RED ID and follower count for screening, but RED ID is
display metadata only. Do not extract Xiaohongshu RED ID values from bios or
build `/user/profile/<RED ID>` URLs.

### Get Creator Profile

```bash
~/.lingzao/bin/lingzao get-user-info --url "https://www.xiaohongshu.com/user/profile/..."
~/.lingzao/bin/lingzao get-user-info --platform xhs --user-id "63c21e0f000000002801a1bb"
~/.lingzao/bin/lingzao get-user-info --platform douyin --user-id "MS4wLjABAAAA..."
~/.lingzao/bin/lingzao get-user-info --platform youtube --user-id "UC..."
~/.lingzao/bin/lingzao get-user-info --url "https://www.tiktok.com/@creator"
~/.lingzao/bin/lingzao get-user-info --url "https://www.instagram.com/creator/"
```

Use this when the user provides a creator profile URL or platform user ID and needs full profile-level stats. For Douyin bare user IDs, use the profile `sec_user_id`. For YouTube, use a channel ID or `/channel/UC...` URL; if the user only has a handle, call `search-users` first. For basic homepage analysis, prefer `get-user-posted-notes` and avoid calling both commands by default.

### Get Creator Recent Posts

```bash
~/.lingzao/bin/lingzao get-user-posted-notes --url "https://www.xiaohongshu.com/user/profile/..."
~/.lingzao/bin/lingzao get-user-posted-notes --platform xhs --user-id "63c21e0f000000002801a1bb"
~/.lingzao/bin/lingzao get-user-posted-notes --platform douyin --user-id "MS4wLjABAAAA..." --limit 20
~/.lingzao/bin/lingzao get-user-posted-notes --platform youtube --user-id "UC..." --limit 20
~/.lingzao/bin/lingzao get-user-posted-notes --platform tiktok --user-id "<search-users returned id>" --limit 20
~/.lingzao/bin/lingzao get-user-posted-notes --platform tiktok --user-id "<search-users returned id>" --cursor "next_cursor_from_previous_response"
~/.lingzao/bin/lingzao get-user-posted-notes --platform instagram --user-id "<search-users returned id>" --limit 20
~/.lingzao/bin/lingzao get-user-posted-notes --platform instagram --user-id "<search-users returned id>" --cursor "next_cursor_from_previous_response"
```

Use this when the user wants to understand what a creator has posted recently. Use this by default for basic creator homepage analysis. Douyin, TikTok, Instagram, and YouTube support `--limit 20` at most per public call. YouTube reads the Videos list only and does not add a separate Shorts request. If the response has `next_cursor`, reuse it with `--cursor`; for TikTok or Instagram, repeat the same creator URL or ID. If the user asks for full profile-level stats, add `get-user-info`; if the user asks for Xiaohongshu post copy, scripts, captions, or transcript text across recent posts, use `analyze-user-profile` instead. For Douyin transcript text, use `extract-video-copy` on selected video URLs. TikTok, Instagram, and YouTube V1 do not support `analyze-user-profile`.

### Analyze Creator Profile

```bash
~/.lingzao/bin/lingzao analyze-user-profile --url "https://www.xiaohongshu.com/user/profile/..." --limit 20
~/.lingzao/bin/lingzao analyze-user-profile --platform xhs --user-id "63c21e0f000000002801a1bb" --limit 40
~/.lingzao/bin/lingzao analyze-user-profile --platform douyin --user-id "MS4wLjABAAAA..." --limit 20
```

Use this when the user wants deeper creator profile data, including post text, covers, commercial signals, and profile-level content signals. For Xiaohongshu, it also includes subtitle/script previews. For Douyin, it does not extract homepage subtitles or transcript text; use `extract-video-copy` on selected video URLs when the user needs spoken copy.
Use `--limit 20` by default. The default Markdown output shows readable subtitle previews when the platform provides them.
Short-window repeats with the same request parameters may reuse the recent successful result without spending credits again; the CLI output will show a no-charge reuse notice. Use `--force-new` only when the user explicitly needs a fresh paid run, and do not loop it: repeated forced refreshes in the short protection window may be rejected with no charge.
If Douyin profile insight sections are temporarily unavailable, the API and CLI can show `partial_data`, `warnings`, or `unavailable_sections`. Explain that homepage works data still returned successfully, and do not treat the missing insight section as proof that there is no data.

Important for Xiaohongshu: the complete profile subtitle/copy Markdown artifact is a top-level response field, not a per-note subtitle URL. Always check:

`data.artifacts.subtitle_markdown.status`
`data.artifacts.subtitle_markdown.url`

Do not search only inside `items[]`. If `data.artifacts.subtitle_markdown.status == "ready"` and `url` exists, download it before deep script or subtitle analysis:

```bash
curl -L "$subtitle_markdown_url" -o /tmp/lingzao-profile-subtitles.md
```

Use the downloaded Markdown file for complete subtitle/copy analysis. Use `--format json` when the user needs the structured fields. JSON includes `data.artifacts.subtitle_markdown.url` for the complete Markdown file when available, and inline `items[].text.subtitle.content/plain_text` are preview-sized to keep the response readable. If the artifact is unavailable, use the inline subtitle fields. For Douyin, expect `data.artifacts.subtitle_markdown.status == "unsupported"` and use the returned profile insights plus selected-video extraction instead.

### Get Post Detail

```bash
~/.lingzao/bin/lingzao get-note-detail --url "https://www.xiaohongshu.com/explore/..." --xhs-note-type image
~/.lingzao/bin/lingzao get-note-detail --platform xhs --note-id "69690331000000001a02266a" --xhs-note-type video
~/.lingzao/bin/lingzao get-note-detail --platform douyin --note-id "7372484715782352169"
~/.lingzao/bin/lingzao get-note-detail --url "https://www.youtube.com/watch?v=..." --content-type video
~/.lingzao/bin/lingzao get-note-detail --platform youtube --note-id "..." --content-type short
~/.lingzao/bin/lingzao get-note-detail --url "https://www.youtube.com/shorts/..."
~/.lingzao/bin/lingzao get-note-detail --url "https://www.tiktok.com/@creator/video/7349541381817355521"
~/.lingzao/bin/lingzao get-note-detail --url "https://www.instagram.com/reel/<code>/"
~/.lingzao/bin/lingzao get-note-detail --platform instagram --note-id "<decimal media id>"
```

The `/shorts/` URL form preserves Short type automatically. For a bare ID,
`watch?v=` URL, or `youtu.be/` URL, pass the `content_type` returned by search as
`--content-type video|short`; Lingzao does not guess type from duration.
YouTube channel/profile URLs are not content-detail inputs. Use
`get-user-info` or `get-user-posted-notes`; for `@handle`, `/c/`, or `/user/`
URLs, use `search-users` first to obtain the canonical channel ID.

Use this when the user asks to analyze one public post.
For Xiaohongshu details, pass `--xhs-note-type image` for 图文 and
`--xhs-note-type video` for 视频. If the note came from `search-notes`,
`get-user-posted-notes`, or `analyze-user-profile`, reuse that item's
`xhs_note_type` value. If detail returns `NOTE_NOT_FOUND_OR_INACCESSIBLE`,
do not switch `--xhs-note-type` and retry automatically; confirm the source
item type or ask the user.

### Get Post Comments

```bash
~/.lingzao/bin/lingzao get-note-comments --url "https://www.xiaohongshu.com/explore/..."
~/.lingzao/bin/lingzao get-note-comments --url "https://www.xiaohongshu.com/explore/..." --sort most_liked
~/.lingzao/bin/lingzao get-note-comments --platform xhs --note-id "69690331000000001a02266a"
~/.lingzao/bin/lingzao get-note-comments --platform douyin --note-id "7372484715782352169"
~/.lingzao/bin/lingzao get-note-comments --platform tiktok --note-id "7349541381817355521" --limit 20
~/.lingzao/bin/lingzao get-note-comments --url "https://www.instagram.com/p/<code>/" --limit 20
~/.lingzao/bin/lingzao get-note-comments --platform instagram --note-id "<shortcode>" --cursor "next_cursor_from_previous_response"
~/.lingzao/bin/lingzao get-note-comments --url "https://www.douyin.com/jingxuan?modal_id=..." --cursor "next_cursor_from_previous_response"
~/.lingzao/bin/lingzao get-note-comments --url "https://youtu.be/..." --sort most_liked --limit 20
```

Use this when the user asks for public comments on one post. The first version returns top-level comments only. Use `--sort most_liked` for Xiaohongshu or YouTube liked-count sorting; Douyin, TikTok, and Instagram support only `latest`, with TikTok using service-default order. If the response has `data.page.next_cursor`, pass that opaque value unchanged with `--cursor` to fetch one next page. For TikTok or Instagram, repeat the same content URL or ID; for YouTube, repeat the same `--sort` with every cursor request.
Before calling Xiaohongshu comments, ask whether the user wants latest comments
or liked-count sorting. For Douyin, TikTok, and Instagram comments, use only `--sort latest`;
do not pass `--sort most_liked`.

### Get WeChat Official-Account Articles

```bash
~/.lingzao/bin/lingzao get-article-detail --url "https://mp.weixin.qq.com/s/..."
~/.lingzao/bin/lingzao get-article-detail --url "https://mp.weixin.qq.com/s/..." --output /tmp/article.md
~/.lingzao/bin/lingzao get-article-stats --url "https://mp.weixin.qq.com/s/..."
~/.lingzao/bin/lingzao get-related-articles --url "https://mp.weixin.qq.com/s/..."
```

Use these when the user provides a public WeChat official-account article URL
and asks to analyze the article, inspect public engagement metrics, or expand
from that article to related public articles. The first version is URL-only and
costs 20 credits per call. An empty related-articles list is a valid response.
Do not use these commands for account article history, account listing, or
multi-page fanout unless Lingzao adds a separate capability.

For full article analysis, prefer `get-article-detail --output /tmp/article.md`.
The command saves the complete article text as a local Markdown file and prints
only the file path plus a short summary in chat. Read the saved Markdown file
for detailed analysis instead of asking the CLI to paste the full article body
into the conversation.

### Extract Short-Video Copy

```bash
~/.lingzao/bin/lingzao extract-video-copy --url "https://www.xiaohongshu.com/explore/..."
~/.lingzao/bin/lingzao extract-video-copy --url "https://v.douyin.com/..."
```

Use this when the user asks for short-video spoken copy, transcript, subtitles, or口播文案.

### Generate Image

```bash
~/.lingzao/bin/lingzao generate-image --prompt "一张小红书封面图，主题是 AI 生图新手避坑，干净明亮，中文大标题留白" --output /tmp/lingzao-image.png
~/.lingzao/bin/lingzao generate-image --prompt "极简产品海报，白底，柔和阴影" --size 1024x1536 --output /tmp/poster.png
~/.lingzao/bin/lingzao generate-image --prompt "参考两张图，保留人物风格，把产品界面换成灵造首页截图" --size 1536x2048 --image /tmp/style.png --image /tmp/product.png --output /tmp/poster.png
~/.lingzao/bin/lingzao generate-image --prompt "批量生成 3 张封面草稿" --count 3 --size 1024x1536 --output /tmp/poster.png
~/.lingzao/bin/lingzao generate-image --prompt-file /tmp/lingzao-prompt.txt --output /tmp/poster.png
```

Use this only when the user asks to generate a creator image asset. For normal
research, do not call image generation automatically.

When the user wants N images from the same prompt, call `generate-image` once
with `--count N` for N=2..5. Do not loop the same prompt as multiple
`--count 1` calls: Lingzao treats short-window identical requests, including
same-count retries, as duplicate POST recovery and returns the same batch
instead of creating new images. If the user wants distinct concepts, vary the
prompt for each concept or use one counted batch for same-prompt variants.

Before calling `generate-image`, run the minimal intake gate. If the user only
says something like "给我做一张某某海报图" or provides only a broad topic, do
not spend credits immediately. Ask for the two visual anchors first:

1. 你有没有参考图？可以发 1-3 张你喜欢的封面/海报/图文截图。
2. 你有没有想要的配色？比如明亮白底、绿色清爽、黑金高级、蓝色科技感。

If those are still unclear, ask at most one extra route-changing question, such
as the publishing platform/size, exact on-image text, or whether the user wants
people/no people. Only proceed directly without asking when the user already
provided enough constraints: topic + platform/format + visual style/reference
or color + on-image text/material.
Use `--image` for local reference images; repeat it for multiple images. The
Skill uploads those files directly to Lingzao for the current request, so the
user does not need to upload them elsewhere first. Supported reference image
formats are png, jpeg, and webp.
For long, Chinese, or multiline prompts, prefer writing the prompt to a UTF-8
text file and passing `--prompt-file /path/to/prompt.txt`, or pipe the prompt
with `--prompt-stdin`, to avoid shell quoting or command-line encoding issues.

#### Reference Image Handling

For Codex, WorkBuddy, and other agent runtimes:

- `--image` accepts local filesystem paths only. If the user provides a
  reference image through a chat attachment, pasted image, screenshot, or input
  box, first materialize that image as a local file before calling the CLI.
  Preserve the original supported image format when saving the file.
- Use a per-run temporary directory for runtime-provided images, for example
  `/tmp/lingzao-image-inputs/<run-id>/ref-1.png` and
  `/tmp/lingzao-image-inputs/<run-id>/ref-2.png`. Use absolute paths in the CLI
  call.
- If the user already provided a stable local path, such as a file under
  `/Users/...`, you may pass that path directly. If the runtime-provided image
  lives in a temporary attachment path, copy it into the per-run temp directory
  first.
- Do not proactively convert image formats. If the input image is already png,
  jpeg, or webp and its file size is reasonable, pass it as-is. Do not convert
  png to webp or jpeg just because an example path uses a different extension.
- Only when a reference image is larger than 2 MB, create a smaller copy in the
  temp directory and pass that copy with `--image`. Keep the file extension and
  actual image bytes consistent. If resizing or compression fails, use the
  original supported image file instead of trying another format.
- Do not overwrite the user's original image file. Do not store reference
  images in the repo. If the runtime cannot save an uploaded or pasted image to
  a local path, ask the user to save the image locally and provide the path.
- In the prompt, state what should be borrowed from the reference images, such
  as layout, color palette, product shape, character style, or composition. Do
  not say only "reference this image" when a more specific instruction is
  possible.

Example with a runtime-provided reference image:

```bash
mkdir -p /tmp/lingzao-image-inputs/run-001 /tmp/lingzao-image-outputs/run-001
~/.lingzao/bin/lingzao generate-image \
  --prompt "参考这张图的排版和明亮色彩，生成一张小红书封面图，主题是 AI 生图新手避坑，中文大标题留白" \
  --size 1024x1024 \
  --image /tmp/lingzao-image-inputs/run-001/ref-1.png \
  --output /tmp/lingzao-image-outputs/run-001/result.png
```

The command creates a Lingzao async batch and automatically polls the returned
status URL until the background job finishes or the command timeout is reached.
Image generation can take several minutes; `--timeout` can extend waiting for
large or slow batches, but does not shorten the built-in per-image polling
window. For one image, `--output` writes the result to the exact path you
provide. For `--count` greater than 1, `--output /tmp/poster.png` writes every
successful image as numbered files such as `/tmp/poster-1.png`,
`/tmp/poster-2.png`, and so on. Default Markdown output requires `--output` so
paid generated images are saved locally. If a direct API caller receives
`GENERATION_IN_PROGRESS` with a returned `poll_url`, use it to poll the active
batch instead of POSTing again. If no `poll_url` is returned, wait briefly and
retry. Use `--format json` only when you need structured automation data.

## Usage Notes

- For profile and post URLs, pass the URL directly when possible.
- For direct IDs, include `--platform`. For Xiaohongshu follow-up profile checks,
  prefer the 24-character `users[].id` returned by `search-users`; RED ID is
  display metadata only.
- Omit `--limit` unless the user asks for a specific count.
- Search notes default to comprehensive sorting, all note types, and all time; use `--sort`, `--note-type`, and `--time-filter` when the user asks for ranked or filtered note search.
- Use `--format json` only when another tool needs structured output.
- Default output is Markdown for agents to read and summarize.
- If the API key or account needs attention, ask the user to open the Lingzao dashboard.