feishu-doc-reader

What This Skill Does

The feishu-doc-reader is a comprehensive toolkit for OpenClaw users to interface directly with the Feishu (Lark) ecosystem. By utilizing the official Feishu Open API, this skill allows users to extract, parse, and analyze content from virtually any document type stored in a Feishu workspace. It abstracts complex API interactions into simple command-line scripts, enabling users to read new-version documents (docx), electronic spreadsheets (sheets), multi-dimensional databases (bitables), and hierarchical knowledge base structures (wiki). Whether you are performing document archival, data extraction for analysis, or automating report generation, this tool provides a robust, standardized bridge between your AI agent and your organizational knowledge base.

Installation

To install this skill, use the ClawHub command: clawhub install openclaw/skills/skills/alfredxia-ai/feishu-document-reader. After installation, follow the configuration steps: Create a file at ./reference/feishu_config.json containing your Feishu app_id and app_secret. Ensure that your script permissions are updated using chmod +x scripts/*.sh to enable execution. For security best practices, ensure the configuration file is protected with chmod 600 and is strictly excluded from any public or version control systems.

Use Cases

1. Automated Knowledge Synthesis: Automatically index and summarize entire Wiki spaces to create searchable local knowledge bases.
2. Data Pipeline Integration: Extract data from Bitable entities directly into Python scripts for further statistical processing or visualization.
3. Content Migration: Convert various Feishu document formats into clean, structured text output for archival or integration with other markdown-based AI workflows.
4. Batch Processing: Use the recursive Wiki scanning feature to monitor project progress updates stored in nested document structures.

Example Prompts

1. "Read the content of the document located at https://xxx.feishu.cn/docx/xxxxx and summarize the key action items from the text."
2. "Extract all records from the Bitable with token 'basexxxxxxxxxxxxxx' and format the output as a clean table."
3. "Recursively scan the Wiki space 'SPACE_ID' and generate a summary report of all updates made this week."

Tips & Limitations

• Always prioritize using the unified read_feishu.sh script, as it handles auto-detection of document types, reducing the risk of errors.
• For complex documents, use the --pretty flag to ensure the output JSON is readable and structured.
• Note that while docx, sheet, bitable, and wiki are fully supported, slides support is currently limited to metadata extraction only.
• Always ensure your Feishu App has the necessary 'read' scopes enabled in the Feishu Developer Console; otherwise, the API will return permission errors even with valid credentials.