The user wants to batch extract multiple Xiaohongshu posts. Please proceed as follows:
Constant definitions
- Cookies file:
~/cookies.json - Obsidian save directory:
~/Documents/Obsidian Vault/xhs - Whisper model:
mlx-community/whisper-large-v3-turbo
Input
List of links provided by the user: $ARGUMENTS
Process
Step 1: Parse links
Extract all Xiaohongshu links from the input (supports multiple lines, space-separated, comma-separated). For each link, extract the post ID and xsec_token.
Step 2: Check Cookies
Check whether ~/cookies.json exists. If it does not exist, follow /xhs step 0 to guide the user through exporting it.
Step 3: Extract one by one
For each link, execute /xhs full extraction process (steps 2-4):
- Request the page → parse INITIAL_STATE
- For video posts, perform speech transcription
- Organize in Peter Thiel style
- Save as
{YYYY-MM-DD} {短标题}.md
Leave a 3-second interval between each post to avoid triggering anti-scraping measures.
Step 4: Summary report
After everything is completed, output a brief summary:
- Number of successes/failures
- Each post's filename and one-sentence summary