feishu-sheet-links

Extracts all hyperlinks from every sheet tab of a public Feishu spreadsheet, then optionally batch-downloads the linked articles as Markdown files.

Workflow

When invoked, follow these steps:

Step 1 — Get the URL

If the user has already provided a Feishu URL, use it. Otherwise ask:

"请提供飞书多维表格的链接（公开可访问的）"

Confirm the URL looks like https://*.feishu.cn/wiki/... or https://*.feishu.cn/sheets/....

Step 2 — Extract links

Resolve the skill directory:

SKILL_DIR="$(find ~/.claude/skills /workspace/.claude/skills -maxdepth 1 -name feishu-sheet-links -type d 2>/dev/null | head -1)"

Run the extraction script:

npx -y bun "${SKILL_DIR}/scripts/main.ts" "<spreadsheet-url>" -o feishu-links.json

Each sheet tab takes 8–15 seconds to load — let the user know it may take a moment.

Step 3 — Show a summary

After extraction, show the user a summary:

• How many sheets were found
• How many links per sheet (with sheet names)
• Total link count

Example:

Found 4 sheets, 127 links total:
- 1月: 32 links
- 2月: 28 links
- 3月: 35 links
- 4月: 32 links

Saved to: feishu-links.json

Step 4 — Offer to download articles

Ask the user if they want to download the linked articles as Markdown:

"是否需要批量下载这些文章为 Markdown 文件？"

If yes, ask for an output directory (default: ./feishu-articles), then run:

npx -y bun "${SKILL_DIR}/scripts/download.ts" feishu-links.json \
  -o <output-dir> \
  -c 5 \
  --max-wait 20000

Download supports resume — if interrupted, re-running skips already-downloaded files.

Error Handling

How It Works

1. Reuses an existing Chrome instance if available (ports 64023, 9222, 9229), otherwise launches its own isolated instance
2. Opens the spreadsheet to discover all sheet IDs from spreadApp.collaborativeSpread._spread.sheetIdToIndexMap
3. For each sheet, opens a dedicated tab at ?sheet=<id>, calls setActiveSheetIndex() to trigger lazy loading, and waits for sheet._dataModel.contentModel to populate
4. Extracts links from two Feishu storage formats:
   • url-type — contentModel.link.idToRef._map (whole-cell hyperlinks)
   • mention-type — contentModel.segmentModel.table (inline rich-text links)

Environment Variables