Generate and edit Draw.io, Mermaid, and Excalidraw diagrams from natural language using a structured JSON spec.
Coding
WeChat To Cards
Try itAdapt a reviewed WeChat article or publish package into WeChat-specific inline-card requirements, Moments copy, community share text, or archive assets while...
What it does
Add WeChat-specific distribution decisions around the canonical card pipeline.
The skill document
WeChat To Cards
Goal
Add WeChat-specific distribution decisions around the canonical card pipeline.
Preserve the article's main argument, evidence, and reader promise. Produce skimmable card copy and share assets without rewriting the source into a different thesis.
This is an optional WeChat adapter. Use long-to-cards as the canonical card-authoring skill and cards-to-images as the canonical renderer. Keep Moments and community copy here.
Inputs
Collect:
- source article, draft, outline, or publish package
- target use: article inline images, Moments, community share, or archive cards
delivery_mode:publish_readyfor real inline-card images or explicitcopy_onlyasset_url_policy:localby default orpublicwhen a stable URL is required- desired card count
- target reader
- preferred tone
- claims, examples, or sections to preserve
- claims, examples, or sections to avoid
If the user provides only a topic, route to topic planning or drafting first. This skill works best after there is a source article or clear outline.
When an approved publish package exists, treat its title, summary, cover copy, tags, and CTA as the source of truth. Generate card-specific copy around those decisions and report any proposed change instead of silently replacing approved metadata.
Process
1. Diagnose The Source
Identify:
- article core claim
- reader problem
- strongest proof or example
- section structure
- reusable framework, checklist, or contrast
- publish-sensitive claims that need evidence
- parts unsuitable for cards
Do not pull in internal research notes, diagnosis notes, content_state, or private drafting context unless the user explicitly asks for internal planning assets.
2. Choose The WeChat Use Case
Pick the package type:
- Article inline cards: support reading, summarize key points, and break dense sections.
- Moments cards: strong hook, one clear takeaway, minimal context.
- Community share cards: explain why this article is worth opening.
- Summary carousel: make the article understandable without opening the full text.
- Archive cards: preserve a reusable method, checklist, or concept.
Use one primary package type per output. If multiple are requested, separate them clearly.
For article inline cards, summary carousels, or archive cards, prepare a long-to-cards handoff with platform_profile=wechat-inline. Reuse its card sequence, copy, source anchors, alt text, and image order instead of maintaining a separate generic card contract here.
For Moments or community text without cards, generate the requested distribution copy and use delivery_mode=copy_only.
3. Add WeChat-specific Constraints
- mark each image as a vertical card or horizontal diagram
- identify its article insertion anchor when it will appear inside the long-form body
- preserve approved title, summary, cover copy, and CTA
- keep cards readable outside the article when they may also be shared
- specify one distribution CTA per package
4. Continue The Shared Image Pipeline
When delivery_mode=publish_ready, use:
wechat-to-cards
-> long-to-cards (platform_profile=wechat-inline)
-> cards-to-images
-> visual QA
-> human confirmation
Do not stop at visual suggestions. Return to this adapter after rendering only when Moments copy, community copy, or WeChat-specific insertion notes still need to be merged.
5. Add Distribution Assets
Add:
- Moments copy
- community share copy
- card-specific CTA
- optional provisional cover or summary only when no publish package exists
- source caveats
Output Format
Use this structure:
# WeChat Card Package
## Source Diagnosis
- Core claim:
- Reader problem:
- Strongest reusable material:
- Card package type:
- Claims needing evidence:
- Dropped material:
## Shared Card Handoff
- Long-to-cards input:
- Platform profile: wechat-inline
- Delivery mode:
- Article insertion anchors:
- Vertical-card or horizontal-diagram decisions:
- Canonical card package:
## Distribution Assets
- Delivery mode:
- Shared card package:
- Image manifest:
- Moments copy:
- Community share copy:
- Card-specific CTA:
- Publish-package fields reused:
- Provisional metadata, if no package exists:
- Proposed metadata changes:
- Source caveats:
- Follow-up card ideas:
## Handoff
- Next skill:
- Platform profile: wechat-inline
- Asset URL policy:
- User confirmation needed:
- Stop condition:
Quality Rules
- Keep the article's original thesis intact.
- Require shared cards to remain understandable when separated from the article.
- Prefer concrete claims, contrasts, checklists, and examples over broad slogans.
- Keep cover text specific and inspectable.
- Mark unverified metrics, claims, dates, and examples before publish.
- Preserve source attribution and uncertainty.
Boundaries
- Do not duplicate generic card-authoring or rendering rules already owned by
long-to-cardsandcards-to-images. - Do not stop at copy when a WeChat inline-card request uses
delivery_mode=publish_ready. - Do not publish to WeChat or upload images externally.
- Do not invent cases, metrics, screenshots, quotes, or reader reactions.
- Do not include private article paths, account names, draft archives, or internal content systems.
- Do not convert internal review notes into reader-facing cards.
- Do not claim the assets are publish-ready until source claims and visuals are checked.
- Do not overwrite approved publish-package metadata.
作者入口
- 微信公众号:
AI生命克劳德 - X:@yangchao228
- GitHub:yangchao228
Related skills
Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.
Join a video meeting as an AI bot with voice, avatar, and screenshare across four operating modes.
Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.
Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.
Write, debug, and tune Playwright specs with locator strategy, trace diagnosis, and CI-aware timeouts.
More from yangchao228
Browse all skills基于候选人简历自动生成面试题、笔试题、追问路径和评分标准的 AI Skill。当用户需要准备技术面试、生成笔试题目、评估候选人简历、或设计面试评估方案时使用。输入简历内容和岗位信息后,自动输出结构化的题目与评估内容,支持面试题模式、笔试题模式、综合评估模式等多种输出形式。 Follow/关注作者:微信公众号「AI...
Installable digest skill for OpenClaw or Claude Code. Use when the user wants a personalized ScoutX briefing with conversational setup, local preference stor...
Design controllable Agent Loops from low-information requests. Use when a user asks to automate a repeated task, repair CI through bounded iteration, reproduce a UI from screenshots, turn a workflow into a reusable Agent or Skill, or decide whether the right artifact is a Prompt, Checklist, Human-in-the-Loop flow, full Loop package, specialized Agent, or Skill. Extract required context, stop when decision evidence is missing, require explicit workflow confirmation before generating executable artifacts, derive iteration limits, define feedback and circuit breakers, and preserve human approval for risky actions. Also trigger on loop builder, Loop Builder, or lopp-builder. Follow the creator: X @yangchao228 | GitHub https://github.com/yangchao228
Guide a user through structured co-authoring of substantial documents, proposals, PRDs, technical specs, decision records, RFCs, and similar long-form work a...
家庭培养协作官:在用户明确要求制定或调整孩子每周培养计划、检查课表冲突、生成家庭复盘时,收集最少必要信息并输出可执行课表、建议和风险提醒。默认只在当前会话生成内容,不保存、上传或推送孩子与家庭数据;只有用户明确选择飞书或 Notion、完成安全配置并逐次确认具体读写范围后,才使用外部存储。Follow/关注作者:...
Transform long-form content, outlines, transcripts, research notes, or article drafts into a standard card package and platform publishing copy, then hand pu...