One command completes it: break down words → forge infographics. Supports multiple words in parallel.
Mode
Force NATIVE mode. This workflow is a pure skill pipeline (ljg-word → ljg-card -i), and does not require the Algorithm's seven-step process. Directly call the skill according to the execution steps below; do not go through OBSERVE/THINK/PLAN/BUILD/EXECUTE/VERIFY/LEARN.
Arguments
Directly pass in one or more English words, separated by spaces.
/ljg-word-flow Obstacle
/ljg-word-flow Serendipity Resilience Entropy
Execution
1. Collect the word list
Extract all English words from the user message.
2. Process each word
For each word, execute two steps in sequence:
Step A — Break down the word (ljg-word):
Call the Skill tool to execute ljg-word, passing in the word. Output the Markdown breakdown result in the conversation.
Step B — Forge infographic (ljg-card -i):
Using the breakdown content from Step A as input, call the Skill tool to execute ljg-card -i. Generate the PNG file to ~/Downloads/。
3. Multiple words in parallel
When there are multiple words, start one Agent subagent per word to process them in parallel (within each subagent, the A→B sequence is internal).
4. Summary report
════ 詞卡完成 ═══════════════════════
📖 {Word1}
🖼️ ~/Downloads/{Word1}.png
📖 {Word2}
🖼️ ~/Downloads/{Word2}.png
...
Key constraints
- Break down the word first, then forge the card; the order is irreversible
- The respective quality standards of ljg-word and ljg-card -i remain unchanged
- The infographic content comes from the word breakdown result, not the dictionary definition