Feishu Doc Search

What This Skill Does

The Feishu Doc Search skill allows the OpenClaw AI agent to seamlessly interface with your Feishu (also known as Lark) Drive environment. By leveraging this tool, the agent can perform full-text searches across your private cloud documents, retrieving relevant file titles, content snippets, and direct access URLs. It acts as an indexing bridge, enabling the agent to synthesize information buried within your organization's internal knowledge base without requiring you to manually navigate the Feishu interface. The skill returns structured JSON data containing tokens and document pointers, facilitating automated workflows that depend on internal corporate documentation.

Installation

To integrate this skill into your local environment, use the OpenClaw CLI tool. Ensure you have the necessary environment variables configured for Feishu API authentication before running the command below:

clawhub install openclaw/skills/skills/autogame-17/feishu-doc-search

Once installed, you can trigger the functionality by invoking the node execution script directly or via the OpenClaw orchestration layer using the provided flag syntax: node skills/feishu-doc-search/index.js --query "your search term here".

Use Cases

This skill is ideal for teams relying on Feishu for project management and documentation. Common use cases include: 1) Automated onboarding assistance where the agent searches for "new employee handbook" to answer starter questions. 2) Project status retrieval by searching for meeting notes from specific dates or project codes. 3) Knowledge management, where users ask the agent to find internal technical architecture diagrams stored in specific folders. 4) Quick retrieval of policy documents to provide instant clarification during team communications.

Example Prompts

1. "Find the latest Feishu document related to the Q4 marketing strategy and summarize the key action items for me."
2. "Search for the technical spec document for the new API module and give me the direct link to the file."
3. "Look through my documents for anything mentioning 'project phoenix' and list all found titles."

Tips & Limitations

The efficiency of this skill depends heavily on your search query precision. Use specific keywords rather than broad terms to reduce noise in the results. Currently, the skill requires read-access permissions to be granted via the Feishu developer console for the specific bot associated with the agent. Note that this skill does not support multi-modal file analysis (like image recognition within documents); it focuses strictly on text-based indexing. For security, ensure your environment variables containing API tokens are stored in a secure secrets manager rather than hardcoded in the skill configuration files.