Ljg Word Flow
Turn one or more English words into deep semantic write-ups plus downloadable infograph PNG cards in a fixed two-step pipeline.
Overview
ljg-word-flow is an agent skill for the Grow phase that runs deep English word analysis then generates a matching infograph PNG card per word.
Install
npx skills add https://github.com/lijigang/ljg-skills --skill ljg-word-flowWhat is this skill?
- Forced NATIVE pipeline: ljg-word then ljg-card -i—skips Algorithm seven-step OBSERVE/THINK/PLAN flow
- Strict order: semantic Markdown analysis before infograph; card content must come from analysis not dictionary gloss
- Multi-word input with optional parallel subagents (each word runs A→B serially)
- Outputs PNG to ~/Downloads/ with a formatted completion summary block
- Triggers on 词卡, word card, word flow, or English words wanting analysis plus visual
- 2-step fixed pipeline: ljg-word → ljg-card -i
- 7-step Algorithm flow explicitly bypassed in NATIVE mode
Adoption & trust: 3.3k installs on skills.sh; 5.5k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You want polished word explainers and shareable infograph cards but manually running analysis and design steps separately is slow and inconsistent.
Who is it for?
Creators who already use ljg-word and ljg-card and want a single slash command for 词卡 / word card workflows.
Skip if: Builders who need only dictionary definitions without visuals, or projects that cannot invoke dependent ljg-word and ljg-card skills.
When should I use this skill?
User says 词卡, word card, word flow, or provides English words wanting both analysis and visual card.
What do I get? / Deliverables
Each requested word yields Markdown deep-dive output in the session and a ~/Downloads/{Word}.png card sourced from that analysis, with a batched completion report for multiple words.
- Per-word Markdown semantic analysis in conversation
- Per-word infograph PNG in ~/Downloads/
Recommended Skills
Journey fit
Vocabulary and explainable content assets support growth through learning material, social posts, and branded visuals rather than core app code. The workflow produces editorial and visual content (Markdown analysis + PNG), not product integrations.
How it compares
Orchestration workflow across two skills—not a standalone analyzer or a generic image generator.
Common Questions / FAQ
Who is ljg-word-flow for?
Solo creators and learners who want English vocabulary deep-dives plus infograph PNGs in one automated pipeline.
When should I use ljg-word-flow?
When the user says 词卡, word card, or word flow, or supplies English words wanting both semantic analysis and a visual card during content or learning prep.
Is ljg-word-flow safe to install?
Check the Security Audits panel on this page; the flow writes PNGs to ~/Downloads/ and depends on sibling skills you should review before use.
SKILL.md
READMESKILL.md - Ljg Word Flow
# ljg-word-flow: 词卡 一条命令完成：解词 → 铸信息图。支持多词并行。 ## 模式 **强制 NATIVE 模式。** 本 workflow 是纯 skill 管道（ljg-word → ljg-card -i），不需要 Algorithm 的七步流程。直接按下方执行步骤调用 skill，不走 OBSERVE/THINK/PLAN/BUILD/EXECUTE/VERIFY/LEARN。 ## 参数 直接传入一个或多个英文单词，空格分隔。 ``` /ljg-word-flow Obstacle /ljg-word-flow Serendipity Resilience Entropy ``` ## 执行 ### 1. 收集单词列表 从用户消息中提取所有英文单词。 ### 2. 处理每个单词 对每个单词，串行执行两步： **步骤 A — 解词（ljg-word）：** 调用 Skill tool 执行 `ljg-word`，传入单词。在对话中输出 Markdown 解析结果。 **步骤 B — 铸信息图（ljg-card -i）：** 以步骤 A 的解析内容为输入，调用 Skill tool 执行 `ljg-card -i`。生成 PNG 文件到 `~/Downloads/`。 ### 3. 多词并行 多个单词时，每个单词启动一个 Agent subagent 并行处理（每个 subagent 内部 A→B 串行）。 ### 4. 汇总报告 ``` ════ 词卡完成 ═══════════════════════ 📖 {Word1} 🖼️ ~/Downloads/{Word1}.png 📖 {Word2} 🖼️ ~/Downloads/{Word2}.png ... ``` ## 关键约束 - 先解词后铸卡，顺序不可逆 - ljg-word 和 ljg-card -i 各自的质量标准不变 - 信息图内容来自解词结果，不是字典释义