A sophisticated routing layer that orchestrates multiple extraction engines to provide high-fidelity web search and content retrieval.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install web-intel
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 web-intel 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 Intel?
Web Intel serves as the unified intelligence gateway for the Openclaw Skills ecosystem, acting as a routing layer rather than a standalone tool. It intelligently manages the complexity of the modern web by orchestrating content-extraction scripts, Jina Reader, Firecrawl, and Chrome DevTools Protocol (CDP) proxies. By abstracting these tools, it allows agents to focus on data analysis while the skill handles the technical nuances of URL classification, anti-bot bypass, and session management.
This skill is engineered for efficiency, offering three distinct tiers of operation—fast, standard, and deep—to optimize for both token consumption and latency. Whether you are performing high-volume market research or deep-diving into authenticated corporate portals, Web Intel ensures you get the cleanest data at the lowest possible cost within the Openclaw Skills framework.
Web Intel Use Cases
- Executing rapid market scans to gather search snippets and titles for trend analysis.
- Extracting full-text content from complex financial news sites like East Money or Xueqiu.
- Performing competitive research within brand marketing workflows to analyze product feedback.
- Accessing and scraping data from websites protected by sophisticated anti-bot measures or requiring login states.
- Delegating specialized extraction tasks for platforms like WeChat, Feishu, and YouTube to specific handlers.
How Web Intel Works
- Receives an input trigger, which can be either a keyword-based search query or a specific target URL.
- Analyzes the URL via the extract_router to determine if specialized platform handlers (e.g., social media or internal docs) are required.
- Selects the optimal toolchain based on the user-defined mode: Jina for lightweight text, Firecrawl for broader searches, or CDP for interactive sessions.
- Initiates the retrieval process, automatically escalating to more advanced tools like web-access CDP if simple extraction fails.
- Standardizes the raw output into a clean JSON format, including metadata such as latency and tool attribution, for consumption by other agents.
Web Intel Setup
To integrate Web Intel with your Openclaw Skills workspace, ensure all upstream dependencies for CDP and search providers are configured.
# Check and install dependencies for the deep-mode CDP proxy
bash ~/.openclaw/workspace/skills/web-access/scripts/check-deps.sh
# Run a test search query using the fast mode
python3 ~/.openclaw/workspace/skills/web-intel/scripts/web_intel.py --query "Openclaw Skills tutorial" --mode fast
Web Intel Data Schema & Taxonomy
Web Intel provides a highly structured output to ensure interoperability between different Openclaw Skills.
| Property | Type | Description |
|---|---|---|
query |
String | The original search term or URL processed. |
mode |
String | The operational tier (fast, standard, or deep). |
tool_used |
String | The specific engine used (e.g., jina, firecrawl_search). |
results |
Array | A list of search results including titles, URLs, and snippets. |
full_content |
String | The complete extracted page content (null in fast mode). |
web_access_available |
Boolean | Status of the CDP Proxy connection. |
latency_ms |
Integer | Total processing time in milliseconds. |
Web Intel Advanced Features
- Intelligent Tiering: Automatically balances speed and cost across three operational modes.
- Seamless CDP Integration: Leverages local Chrome instances via localhost:3456 to maintain login states and session persistence.
- Recursive Delegation: Automatically hands off requests to specialized handlers for non-standard web content like YouTube transcripts or Feishu documents.
- Multi-Agent Orchestration: Directly compatible with Brand Marketing Workflows (BMW) and financial research sub-agents.
- Failover Logic: Automatically downgrades or upgrades the extraction method if the primary tool encounters anti-bot restrictions.
SKILL.md
Loading
Related Openclaw Skills
Facebook Manager v2
A comprehensive Facebook management tool for Openclaw that automates Fanpage engagement and personal account data retrieval using Meta Graph APIs.
Browser Search Ultimate CN
A lightweight Python skill that uses system-level curl requests to fetch and parse domestic Chinese search results directly from engines like Bing.
Loki Log Query
A powerful utility for querying Grafana Loki logs via API to streamline debugging and Kubernetes troubleshooting.
Regression Test Generator
An intelligent test case recommendation engine that automatically identifies and prioritizes regression tests based on software change impact.
zzheng Legal Article Generator
A specialized automation tool for legal professionals to generate trending WeChat Official Account articles from real-time news sources.
BlockBeats API Skill
A comprehensive crypto market intelligence skill providing real-time news, institutional ETF flows, on-chain analytics, and global macro indicators.
Featured*
