AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Feishu (Lark) Doc Scraper

Extracts Feishu/Lark docs, wiki collections, sheets and Minutes transcripts into faithful local Markdown via the lark-cli API.

Docs & OfficeAdvanced★ 1,323⑂ 212AI score 8/10Last updated: Aug 8, 2026

What it does

  • Classifies any Feishu/Lark URL (wiki node, docx, sheets, Minutes) and pulls the body programmatically through lark-cli, writing it to disk via jq/scripts so the model never paraphrases source text.
  • Treats a collection/hub as a reference graph: extracts <mention-doc>, <sheet>, <image>, cross-tenant and Minutes links with a bundled Python extractor and recurses breadth-first until every leaf is fetched.
  • Handles permission walls honestly — codes 131006/99991679 are server-side boundaries, so it switches to the owner-exported .docx path and restores heading levels from font sizes plus w:shd highlights.
  • Uses the platform's native Minutes transcription API instead of re-running ASR on downloaded media.
  • Ships a strict acceptance contract: residual rich-media tag grep must be empty, no U+FFFD replacement characters, docx output visually verified from a render, source recorded in frontmatter, and permission/image gaps listed explicitly.

Who it's for

  • Teams on Feishu/Lark who need to migrate internal wikis into Obsidian or another local knowledge base
  • Anyone archiving a multi-document collection without silently dropping references
  • People preserving meeting transcripts with speaker labels and timestamps intact
  • Prerequisites: lark-cli (npm @larksuite/cli) authenticated to the tenant; the docx path also needs python-docx and pandoc

Example uses

  1. "Archive this whole Feishu wiki collection as Markdown" → resolves the node token to a doc token, recurses the reference graph, then greps for unresolved tags before stopping.
  2. "Just get the transcript from this 妙记 link" → device-flow login for the extra scope, then the native transcript endpoint.
  3. "This doc says permission denied" → confirms 131006, asks the owner for a .docx export, and converts it with heading/highlight restoration plus a visual diff.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/daymade/claude-code-skills/HEAD/feishu-doc-scraper/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the feishu-doc-scraper folder from the GitHub repo daymade/claude-code-skills into my ~/.claude/skills/feishu-doc-scraper/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/daymade/claude-code-skills.git && mkdir -p ~/.claude/skills && cp -r claude-code-skills/feishu-doc-scraper ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and clone the repo: git clone https://github.com/daymade/claude-code-skills.git
  2. Copy the skill into place: mkdir -p ~/.claude/skills && cp -r claude-code-skills/feishu-doc-scraper ~/.claude/skills/
  3. Install the primary CLI: npm i -g @larksuite/cli (verified against 1.0.32)
  4. Authenticate lark-cli against the target Feishu/Lark tenant; for Minutes transcripts, complete the device-flow login to grant the extra scope.
  5. Add the proxy-bypass variable for Feishu domains: export LARK_CLI_NO_PROXY=1 (put it in your shell profile).
  6. If you expect the docx path, also install pip install python-docx and pandoc.
  7. Restart Claude Code and ask something like "extract this Feishu doc to Markdown" with the URL to trigger the skill.
View source on GitHub ↗License: MIT