feishu-doc-reader

What This Skill Does

The feishu-doc-reader skill provides a robust bridge between the OpenClaw agent and your Feishu (Lark) ecosystem. By leveraging the official Feishu Open API, this skill enables the autonomous agent to traverse, read, and extract content from various document formats, including modern docx files, legacy docs, spreadsheets, and slides. It translates the hierarchical, block-based structure of Feishu documents into clean, machine-readable JSON, allowing the AI to understand the relationship between text, tables, images, and nested elements.

Installation

To install the skill, run the command: clawhub install openclaw/skills/skills/snowshadow/feishu-doc-reader from your terminal. After installation, perform the following setup steps:

1. Navigate to ./reference/ and create a feishu_config.json file containing your Feishu app_id and app_secret.
2. Ensure the scripts are executable by running chmod +x scripts/read_doc.sh and chmod +x scripts/get_blocks.sh.
3. Secure your configuration file by applying restrictive permissions (chmod 600) to prevent unauthorized access to your credentials.

Use Cases

• Automated Documentation Parsing: Feed technical documentation or product requirements from Feishu directly into the agent for summary or Q&A.
• Data Analysis: Extract complex table structures from Feishu Sheets to perform trend analysis or data visualization within the agent's context.
• Content Migration: Transform existing internal knowledge bases into different formats or populate new databases by reading existing docx hierarchies.
• Meeting Follow-up: Extract meeting notes and action items from Feishu docs to sync them with project management tools.

Example Prompts

1. "OpenClaw, read the Feishu document at https://example.feishu.cn/docx/DOC_TOKEN and provide a three-bullet point summary of the action items listed under the 'Q4 Planning' heading."
2. "Extract all the table data from the sheet associated with the Feishu doc token docx_AbCdEfGhIjKlMnOpQrStUv and convert it into a CSV format for my local analysis."
3. "Get the full block hierarchy of the document with token docx_AbCdEfGhIjKlMnOpQrStUv and identify any references to 'Budget' across the entire document structure."

Tips & Limitations

• Structure Sensitivity: Because Feishu documents are block-based, ensure you are using the get_blocks.sh script if you need deep nesting or parent-child context.
• Permissions: Ensure your Feishu app has the necessary read-only permissions in the Feishu Developer Console for the specific documents or spaces you intend to access.
• Security: Never commit your feishu_config.json file to Git; use environment variables or encrypted secrets management where possible to protect your API credentials.
• Slides: Currently, slide extraction is limited to metadata; complex embedded media may require additional API permissions or specific handling within the script.