Thunder Core enables programmatic control over the Windows Thunder (Xunlei) download client via its COM interface to automate and manage complex downloading workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install thunder-core
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 thunder-core 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 Thunder Download Core?
Thunder Download Core is a robust integration designed for Windows users that bridges the gap between Python automation scripts and the popular Thunder (Xunlei) download engine. By leveraging the native Windows COM interface (specifically ThunderAgent.ThunderAgent), this skill allows developers and AI agents to programmatically add download tasks, customize HTTP headers (including Referer, User-Agent, and Cookies), commit download queues, and inspect ongoing tasks.
Integrating this package into Openclaw Skills empowers your AI agents to seamlessly handle large-scale media extraction, batch system updates, and automated asset recovery. This makes it an invaluable utility for automated pipelines that require high-speed, resume-supported downloads directly through the desktop Xunlei engine without manual human interaction.
Thunder Download Core Use Cases
- Batch downloading web assets, media files, or dataset components directly through an AI agent's trigger commands.
- Automating file transfers that require specialized request headers (such as custom User-Agents or anti-leech Referers).
- Incorporating automated download management into system administrative scripts when triggers like 'download' or 'manage downloads' are detected.
- Creating self-operating media scrapers that queue high-speed download tasks to local storage folders.
How Thunder Download Core Works
- Initialization: The skill triggers the Windows COM component
ThunderAgent.ThunderAgentto establish a secure link with the active or background Xunlei instance. - Task Registration: The user queues a target file URL along with destination path and custom request headers (e.g., Referer or cookies).
- Task Submission: The module commits the queued download tasks to the Thunder engine's core processor.
- Task Management: The workflow dynamically tracks status, provides options to query ongoing download lists, or cancels active operations.
- Resource Cleanup: The skill safely releases the COM interface bindings and system resources on completion.
Thunder Download Core Setup
Prerequisites
- A Windows OS environment with the Thunder (迅雷) client installed.
- Python 3.x environment.
Installation
Ensure you have the required pywin32 package installed on your system:
pip install pywin32
Execution
If COM registration fails or permissions are restricted, run your shell or IDE as Administrator. Initialize the skill with the following Python snippet:
from thunder_core import ThunderCore
tc = ThunderCore()
if tc.initialize():
print("Thunder COM Interface initialized successfully!")
Thunder Download Core Data Schema & Taxonomy
The thunder-core skill exposes programmatic methods to feed data parameters directly into the Windows COM bridge. Below is the API reference structure for the core components:
ThunderCore Methods
| Method | Parameter | Description |
|---|---|---|
initialize() |
None | Boots up and registers the underlying COM component. |
add_task(url, save_path, ...) |
url (str), save_path (str), refer_url (str), user_agent (str) |
Registers a download task with customizable connection headers. |
commit_tasks() |
None | Submits all queued tasks to the engine for downloading. |
add_and_commit(...) |
Match add_task args |
Combines the registration and submission process into one step. |
cancel_all() |
None | Stops and removes all ongoing tasks in the session. |
get_tasks() |
None | Returns the list of active tasks. |
close() |
None | Cleanly releases COM bindings. |
Utility Functions
| Function | Parameter | Description |
|---|---|---|
download(url, save_path, ...) |
Match add_task args |
Instant single-file downloader that runs the setup, execution, and teardown in a single step. |
Thunder Download Core Advanced Features
- Granular HTTP Header Control: Customize Cookies, User-Agent strings, and Referer headers to bypass anti-scraping mechanisms and download from restricted endpoints.
- Batch Task Queuing: Register multiple tasks in a queued state and commit them at a precise moment with
commit_tasksto save system resources. - Administrative COM Rights Handling: Built-in compatibility guidelines for handling UAC and administrator privileges to register system-level interfaces cleanly.
- Seamless Openclaw Skills Integration: Easily hooks into natural language triggers like "xunlei download" or "check download progress" to enable multi-agent desktop automation workflow chains.
SKILL.md
Loading
Related Openclaw Skills
Bloom Tutor
An interactive, one-on-one Socratic AI tutoring system designed to achieve 2 Sigma level learning outcomes through adaptive markdown-based lessons.
learn-deep
An advanced learning skill that orchestrates five cognitive perspectives—crossover, occam, graph, prototype, and feynman—to quickly and deeply unpack any concept.
learn-feynman
A structured Feynman technique workflow designed to test your understanding of complex concepts by prompting you to teach them simply to an inquisitive AI student.
learn-prototype
An active learning agent skill that forces users to bypass analysis paralysis, build a raw prototype immediately, and iteratively refine it through guided self-correction.
Learn Crossover
A cognitive scaffolding skill that accelerates the learning of complex concepts by mapping their underlying structures to mental models you already master.
China Happy 8 (KL8) Lottery Data Analyzer
A dedicated Python-based analytical utility for querying historical China Happy 8 (KL8) lottery draw data and statistical frequency.
Featured*
