A powerful real-time web search and image retrieval tool for AI agents integrated with the Feishu communication platform.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dashscope-web-search-feishu
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install dashscope-web-search-feishu using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Web Search for Feishu?
The Web Search for Feishu skill empowers your AI agent to break free from training data limitations by accessing live internet data. Built on the DashScope Qwen API, it provides a seamless bridge between real-time information retrieval and the Feishu ecosystem. Whether your agent needs to track breaking news, verify facts, or fetch visual references, this skill provides the necessary scripts to deliver structured, cited, and high-quality results directly into your workflow.
As part of the growing ecosystem of Openclaw Skills, this tool supports both lightweight turbo searches for quick facts and intensive deep-reasoning tasks for complex research. It is specifically optimized for Feishu, allowing agents to not only retrieve text but also push image assets directly into active chat sessions, making it an essential utility for modern automated workflows.
Web Search for Feishu Use Cases
- Answering questions about current events, stock prices, or weather updates in real-time.
- Performing deep research and fact-checking using multi-source verification for technical reports.
- Fetching and sending relevant images or visual references directly to a Feishu chat upon user request.
- Monitoring recent developments by filtering results based on specific recency windows (e.g., last 7 days).
- Conducting domain-specific searches restricted to trusted websites for professional analysis.
How Web Search for Feishu Works
- The AI agent identifies a need for real-time information or visual content based on the user's natural language input.
- The skill selects the appropriate operation mode: Mode A for text-only results or Mode B for image-centric requests.
- A Python script executes the search query against the DashScope Qwen API, applying requested modifiers like depth, reasoning, or site filtering.
- If in Mode B, the results are piped to a specialized Feishu image script that handles the secure upload and delivery of visual messages.
- The agent synthesizes the search results, preserving academic-style citations for transparency and delivering a clean response to the user.
Web Search for Feishu Setup
To get started with this skill within the Openclaw Skills framework, ensure you have Python 3 installed and the necessary API keys configured in your environment.
# Export required environment variables
export DASHSCOPE_API_KEY="your_key_here"
export FEISHU_APP_ID="your_app_id"
export FEISHU_APP_SECRET="your_app_secret"
# Example: Perform a deep research search
python3 scripts/web_search.py --deep --think "Latest developments in LLM agents"
# Example: Send images to a specific Feishu chat
python3 scripts/web_search.py --images "SpaceX Starship launch" | python3 scripts/feishu_image.py --send --chat-id YOUR_CHAT_ID
Web Search for Feishu Data Schema & Taxonomy
The skill manages data through specific command-line flags and outputs structured text or media. It organizes metadata as follows:
| Feature | Parameter | Data Organization |
|---|---|---|
| Search Mode | --deep / --agent |
Determines the complexity of the retrieval chain. |
| Reasoning | --think |
Prepends a <thinking> block containing the AI's internal logic. |
| Recency | --fresh N |
Restricts the data temporal window to N days. |
| Filtering | --sites |
Constrains the search index to a comma-separated list of domains. |
| Image Handling | feishu_image.py |
Processes image URLs and maps them to Feishu's internal message format. |
Web Search for Feishu Advanced Features
- Multi-agent support using the
--agentflag for complex questions requiring iterative retrieval and synthesis. - Deep reasoning capabilities with the
--thinkmode to provide analysis and trend prediction rather than just raw facts. - Flexible receiver targeting for Feishu, supporting chat IDs, open IDs, and user IDs.
- High-precision domain targeting to ensure your Openclaw Skills only pull information from verified, high-authority sources.
SKILL.md
Loading
Related Openclaw Skills
Best Skill
The premier solution for enhancing AI coding agent workflows with superior logic and efficiency.
OptionsHawk
OptionsHawk provides advanced options market intelligence, including flow analysis, unusual activity detection, and strategy evaluation for equities and ETFs.
Awesome Skills Explorer
A comprehensive discovery engine for finding and installing the best community-driven Openclaw Skills for your AI agent.
Deep Scraper
A high-performance scraping engine that uses Docker and Crawlee to extract clean, structured data from complex websites like YouTube and X/Twitter.
Web Search
A high-performance web search skill that integrates DashScope Qwen to provide real-time information, citations, and multi-modal results for AI agents.
LHON Research Assistant
A specialized research skill designed to coordinate AI-driven discovery and data synthesis to help find a cure for Leber's Hereditary Optic Neuropathy (LHON).
METADATA
Featured*
