Spydra is a high-performance web scraping and data extraction tool designed for researchers and businesses to programmatically collect web information.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install spydra
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 spydra 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 Spydra?
Spydra enables AI agents to interact deeply with web data by managing Organizations, Pipelines, and Users through a structured API. By leveraging Openclaw Skills, developers can integrate sophisticated scraping logic into their AI workflows without manually handling the complexities of browser sessions or raw HTML parsing.
This integration uses the Membrane framework to bridge the gap between AI agents and Spydra's infrastructure. It provides a secure, abstracted layer that allows agents to focus on the data extraction logic while Membrane handles the underlying authentication plumbing and credential refreshing automatically.
Spydra Use Cases
- Automating recurring data collection for market research and competitive analysis.
- Managing complex scraping pipelines and document workflows within an AI-driven workspace.
- Synchronizing extracted web data directly into data science environments or business intelligence tools.
- Orchestrating user and organization-level data permissions via AI agent commands.
How Spydra Works
- The AI agent initiates a connection to the Spydra domain via the Membrane CLI to establish a secure link.
- The system checks for an existing connection; if one is missing, it triggers a client action for the user to authenticate.
- Once the state is READY, the agent searches for specific Spydra actions using natural language intents (e.g., "list pipelines").
- The agent executes the selected action with JSON parameters, which Membrane translates into authenticated API requests.
- If pre-built actions are insufficient, the agent uses the proxy request feature to communicate directly with Spydra endpoints while maintaining managed security headers.
Spydra Setup
First, install the necessary CLI tool globally:
npm install -g @membranehq/cli@latest
Authenticate your agent by specifying the tenant and agent type:
membrane login --tenant --clientName=openclaw
Ensure a connection to the Spydra service is active:
membrane connection ensure "https://www.spydra.app/" --json
Spydra Data Schema & Taxonomy
The Spydra Openclaw Skills integration organizes data around several core entities. The following table describes the primary data objects:
| Entity | Description |
|---|---|
| Project | The top-level container for scraping tasks and data sets. |
| Document | Individual data entries or pages captured during a crawl. |
| User | Account-level metadata and permission scopes. |
| Workspace | Shared environments for team-based data extraction. |
| Action | Definitions of executable functions, including inputSchema and outputSchema. |
Spydra Advanced Features
- Natural Language Action Discovery: Search for capabilities using human-like queries rather than strict endpoint documentation.
- Automatic Credential Management: Full lifecycle handling of OAuth and API tokens with zero local secrets storage.
- Proxy Request Layer: Send custom HTTP methods (POST, PATCH, DELETE) directly to Spydra while retaining managed authentication headers.
- Polling & Long-running Tasks: Built-in support for waiting on building connections or long-running scraping jobs via the
--waitflag.
SKILL.md
Loading
Related Openclaw Skills
Mambo TTS
A high-performance text-to-speech skill for Openclaw providing lively, pre-configured Chinese voice profiles based on Edge TTS.
Skill Creator
Skill Creator is a specialized meta-skill designed to help developers build, test, and refine new AI agent capabilities through an iterative, data-driven workflow.
Workspace Local Retrieval
A professional framework for building local-first retrieval architectures with strict security boundaries and automated maintenance workflows.
Sendoso
A comprehensive integration for managing personalized gifting, eGifts, and direct mail campaigns through AI agents.
Cloudflare Media
A powerful media generation skill leveraging Cloudflare Workers AI for professional text-to-image and text-to-speech capabilities.
AI Earnings Tracker
An automated financial monitoring tool that tracks tech and AI earnings reports, generates AI summaries, and sets alert notifications.
Featured*
