lark-doc

XiaoMi's avatarfrom XiaoMi

Fetch Lark/Feishu document and convert to Markdown. Use this skill to get troubleshooting guides and operation instructions from Lark documents.

⭐1.2kstars🔀179forks📁View on GitHub🕐Updated Dec 26, 2025
[IT Helpdesk and Runbooks]

When & Why to Use This Skill

This Claude skill enables seamless integration with Lark (Feishu) to fetch documents and convert them into Markdown format. It streamlines knowledge management by allowing users to retrieve troubleshooting guides and append resolved incident summaries directly to existing documents, ensuring that operational runbooks are always up-to-date and easily accessible for AI-driven support.

Use Cases

  • Case 1: Retrieving technical SOPs or operation instructions from Lark during a live troubleshooting session to ensure standard procedures are followed.
  • Case 2: Converting complex Feishu documentation into Markdown for easier processing by AI agents or for integration into developer-friendly wikis.
  • Case 3: Automating the documentation of incident resolutions by appending problem descriptions, steps taken, and final solutions to a centralized Lark knowledge base.
  • Case 4: Providing quick access to step-by-step guides for non-technical staff by querying internal enterprise documents through a conversational interface.
namelark-doc
descriptionFetch Lark/Feishu document and convert to Markdown. Use this skill to get troubleshooting guides and operation instructions from Lark documents.

Lark Document Skill

This skill provides two capabilities:

  1. Fetch a Lark (Feishu) document and convert its content to Markdown format
  2. Append troubleshooting summaries to the document for future reference

When to Use

Fetch Document:

  • User needs troubleshooting steps or operation guides
  • Before diagnosing issues, check if there are relevant operation instructions
  • When user asks about specific procedures or step-by-step guides
  • Keywords: troubleshooting, steps, guide, procedure, operation, how to fix

Append Summary:

  • After completing a troubleshooting session
  • When user wants to document a resolved issue for future reference
  • To add new troubleshooting cases to the guide document
  • Keywords: save, record, document, write summary, add to guide

How to Use

Fetch Document

uv run python .claude/skills/lark-doc/fetch_doc.py

Append Troubleshooting Summary

uv run python .claude/skills/lark-doc/append_summary.py \
  -t "问题标题" \
  -p "问题描述" \
  -s '["排查步骤1", "排查步骤2", "排查步骤3"]' \
  -o "解决方案" \
  -n "可选备注"

Parameters:

  • -t, --title: Title of the troubleshooting case (required)
  • -p, --problem: Description of the problem (required)
  • -s, --steps: JSON array of troubleshooting steps taken (required)
  • -o, --solution: The solution that resolved the issue (required)
  • -n, --notes: Optional additional notes

Environment Variables

  • LARK_APP_ID: Lark application ID
  • LARK_APP_SECRET: Lark application secret
  • LARK_DOC_URL: The Lark document URL to fetch/append to
  • LARK_DOMAIN: (Optional) Custom domain for enterprise Lark

Output Format

Fetch Document

Returns Markdown formatted content of the Lark document, including:

  • Headings (H1-H9)
  • Text paragraphs
  • Ordered and unordered lists
  • Code blocks
  • Quotes
  • Todo items (checkboxes)

Append Summary

Returns JSON with success status:

{
  "success": true,
  "documentId": "xxx",
  "url": "https://...",
  "message": "Successfully appended troubleshooting summary: ..."
}

Example Output

Fetched Document

# Troubleshooting Guide

## Common Issues

### Issue 1: Service Not Starting

1. Check the logs
2. Verify configuration
3. Restart the service

### Issue 2: Connection Timeout

- Check network connectivity
- Verify firewall rules

Appended Summary Structure

The summary will be appended to the document with the following structure:

---

### 问题标题 (2025-01-15 14:30)

#### 问题描述

问题的详细描述...

#### 排查步骤

- 排查步骤1
- 排查步骤2
- 排查步骤3

#### 解决方案

解决方案的详细说明...

#### 备注

可选的备注信息...