Agent skill

feishu-doc-reader

Read and extract content from Feishu (Lark) documents using the official Feishu Open API

Stars 129
Forks 41

Install this agent skill to your Project

npx add-skill https://github.com/zephyrwang6/myskill/tree/main/feishu-doc-reader

Metadata

Additional technical details for this skill

moltbot
{
    "emoji": "\ud83d\udcc4",
    "requires": {
        "bins": [
            "python3",
            "curl"
        ]
    }
}

SKILL.md

Feishu Document Reader

This skill enables reading and extracting content from Feishu (Lark) documents using the official Feishu Open API.

Configuration

Set Up the Skill

  1. Create the configuration file at ./reference/feishu_config.json with your Feishu app credentials:
json
{
  "app_id": "your_feishu_app_id_here",
  "app_secret": "your_feishu_app_secret_here"
}
  1. Make sure the scripts are executable:
bash
chmod +x scripts/read_doc.sh
chmod +x scripts/get_blocks.sh

Security Note: The configuration file should be kept secure and not committed to version control. Consider using proper file permissions (chmod 600 ./reference/feishu_config.json).

Usage

Basic Document Reading

To read a Feishu document, you need the document token (found in the URL: https://example.feishu.cn/docx/DOC_TOKEN).

Using the shell script (recommended):

bash
# Make sure environment variables are set first
./scripts/read_doc.sh "your_doc_token_here"

# Or specify document type explicitly
./scripts/read_doc.sh "docx_token" "doc"
./scripts/read_doc.sh "sheet_token" "sheet"

Get Detailed Document Blocks (NEW)

For complete document structure with all blocks, use the dedicated blocks script:

bash
# Get full document blocks structure
./scripts/get_blocks.sh "docx_AbCdEfGhIjKlMnOpQrStUv"

# Get specific block by ID
./scripts/get_blocks.sh "docx_token" "block_id"

# Get blocks with children
./scripts/get_blocks.sh "docx_token" "" "true"

Using Python directly for blocks:

bash
python scripts/get_feishu_doc_blocks.py --doc-token "your_doc_token_here"
python scripts/get_feishu_doc_blocks.py --doc-token "docx_token" --block-id "block_id"
python scripts/get_feishu_doc_blocks.py --doc-token "docx_token" --include-children

Supported Document Types

  • Docx documents (new Feishu docs): Full content extraction with blocks, metadata, and structure
  • Doc documents (legacy): Basic metadata and limited content
  • Sheets: Full spreadsheet data extraction with sheet navigation
  • Slides: Basic metadata (content extraction requires additional permissions)

Features

Enhanced Content Extraction

  • Structured output: Clean JSON with document metadata, content blocks, and hierarchy
  • Complete blocks access: Full access to all document blocks including text, tables, images, headings, lists, etc.
  • Block hierarchy: Proper parent-child relationships between blocks
  • Text extraction: Automatic text extraction from complex block structures
  • Table support: Proper table parsing with row/column structure
  • Image handling: Image URLs and metadata extraction
  • Link resolution: Internal and external link extraction

Block Types Supported

  • text: Plain text and rich text content
  • heading1/2/3: Document headings with proper hierarchy
  • bullet/ordered: List items with nesting support
  • table: Complete table structures with cells and formatting
  • image: Image blocks with tokens and metadata
  • quote: Block quotes
  • code: Code blocks with language detection
  • equation: Mathematical equations
  • divider: Horizontal dividers
  • page: Page breaks (in multi-page documents)

Error Handling & Diagnostics

  • Detailed error messages: Clear explanations for common issues
  • Permission validation: Checks required permissions before making requests
  • Token validation: Validates document tokens before processing
  • Retry logic: Automatic retries for transient network errors
  • Rate limiting: Handles API rate limits gracefully

Security Features

  • Secure credential storage: Supports both environment variables and secure file storage
  • No credential logging: Credentials never appear in logs or output
  • Minimal permissions: Uses only required API permissions
  • Access token caching: Efficient token reuse to minimize API calls

Command Line Options

Main Document Reader

bash
# Python script options
python scripts/read_feishu_doc.py --help

# Shell script usage
./scripts/read_doc.sh <doc_token> [doc|sheet|slide]

Blocks Reader (NEW)

bash
# Get full document blocks
./scripts/get_blocks.sh <doc_token>

# Get specific block
./scripts/get_blocks.sh <doc_token> <block_id>

# Include children blocks
./scripts/get_blocks.sh <doc_token> "" true

# Python options
python scripts/get_feishu_doc_blocks.py --help

API Permissions Required

Your Feishu app needs the following permissions:

  • docx:document:readonly - Read document content
  • doc:document:readonly - Read legacy document content
  • sheets:spreadsheet:readonly - Read spreadsheet content

Error Handling

Common errors and solutions:

  • 403 Forbidden: Check app permissions and document sharing settings
  • 404 Not Found: Verify document token is correct and document exists
  • Token expired: Access tokens are valid for 2 hours, refresh as needed
  • App ID/Secret invalid: Double-check your credentials in Feishu Open Platform
  • Insufficient permissions: Ensure your app has the required API permissions
  • 99991663: Application doesn't have permission to access the document
  • 99991664: Document doesn't exist or has been deleted
  • 99991668: Token expired, need to refresh

Examples

Extract document with full structure

bash
# Read document
./scripts/read_doc.sh "docx_AbCdEfGhIjKlMnOpQrStUv"

Get complete document blocks (NEW)

bash
# Get all blocks with full structure
./scripts/get_blocks.sh "docx_AbCdEfGhIjKlMnOpQrStUv"

# Get specific block details
./scripts/get_blocks.sh "docx_AbCdEfGhIjKlMnOpQrStUv" "blk_xxxxxxxxxxxxxx"

Process spreadsheet data

bash
./scripts/read_doc.sh "sheet_XyZ123AbCdEfGhIj" "sheet"

Extract only text content (Python script)

bash
python scripts/read_feishu_doc.py --doc-token "docx_token" --extract-text-only

Security Notes

  • Never commit credentials: Keep app secrets out of version control
  • Use minimal permissions: Only request permissions your use case requires
  • Secure file permissions: Set proper file permissions on secret files (chmod 600)
  • Environment isolation: Use separate apps for development and production
  • Audit access: Regularly review which documents your app can access

Troubleshooting

Authentication Issues

  1. Verify your App ID and App Secret in Feishu Open Platform
  2. Ensure the app has been published with required permissions
  3. Check that environment variables or config files are properly set
  4. Test with the test_auth.py script to verify credentials

Document Access Issues

  1. Ensure the document is shared with your app or in an accessible space
  2. Verify the document token format (should start with docx_, doc_, or sheet_)
  3. Check if the document requires additional sharing permissions

Network Issues

  1. Ensure your server can reach open.feishu.cn
  2. Check firewall rules if running in restricted environments
  3. The script includes retry logic for transient network failures

Blocks-Specific Issues

  1. Empty blocks response: Document might be empty or have no accessible blocks
  2. Missing block types: Some block types require additional permissions
  3. Incomplete hierarchy: Use --include-children flag for complete block tree

References

Expand your agent's capabilities with these related and highly-rated skills.

zephyrwang6/myskill

start-work

每日工作启动助手。读取Obsidian收件箱、计划文件,提醒今日待办,询问内容创作计划,展示周计划进度,调用热点采集。触发词:"开始工作"、"开启新一天"、"今天做什么"。帮助用户快速进入高效工作状态。

129 41
Explore
zephyrwang6/myskill

doc-coauthoring

Guide users through a structured workflow for co-authoring documentation, articles, or long-form content. Use when user wants to write documentation, proposals, articles, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting articles, or using "co-authoring" workflow.

129 41
Explore
zephyrwang6/myskill

mem-query

AI个人记忆系统的记忆查询功能。检索各层级记忆文件,综合多层级信息回答用户问题。使用场景:(1) 用户问"我的记忆中关于XXX"时;(2) 用户询问自己的习惯/偏好/价值观时;(3) 需要基于用户历史提供建议时。该skill会自动检索L1-L4各层级,引用来源,给出基于记忆的个性化回答。

129 41
Explore
zephyrwang6/myskill

article-review

根据原文内容撰写深度文章评价/解读。当用户提供一篇文章、博客、公众号文章或任何长文内容,并要求生成评价、解读、读后感或二次创作内容时使用此技能。适用于:(1) 对技术文章、行业分析、年终总结等进行深度解读,(2) 提炼文章核心观点并用通俗语言重新表达,(3) 为社交媒体传播生成二次内容。

129 41
Explore
zephyrwang6/myskill

mem-record

AI个人记忆系统的记忆记录功能。自动从对话中提炼关键信息并记录到相应层级。使用场景:(1) 用户说"记录到记忆系统"、"记住这个"、"把这次对话记下来"时;(2) 检测到重要事件、决策、偏好表达时;(3) 用户完成重要任务或做出决策时。该skill会自动判断应该记录到L1情境层、L2行为层、L3认知层,还是建议更新L4核心层。

129 41
Explore
zephyrwang6/myskill

remotion-video

使用 Remotion 框架以编程方式创建视频。Remotion 让你用 React 组件定义视频内容,支持动画、字幕、音乐可视化、3D 视频、教程讲解视频等。适用于程序化视频、批量生成、数据驱动视频、音乐可视化、自动字幕等场景。

129 41
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results