← Back to Skills Marketplace
211
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install opencli-skills
Description
Fetch data from social, video, and news websites using OpenCLI by reusing Chrome login sessions without API keys.
Usage Guidance
This skill is coherent with its stated goal (scraping sites by reusing your Chrome session) but that same mechanism gives it potential access to all data in any logged-in browser session. Before installing: (1) verify the npm package and author (@jackwener/opencli) — review the package source/release on npm/GitHub; (2) review the Browser Bridge extension's permissions and provenance; (3) avoid using this with highly sensitive accounts or run it in a dedicated browser profile with only the accounts you intend to expose; (4) be aware some commands perform write actions (send messages, downloads) and additional tools (yt-dlp) may be required; (5) prefer read-only usage and test in a controlled environment. If the publisher or upstream project is unknown or you cannot audit the code and extension, consider this higher risk and treat it accordingly.
Capability Analysis
Type: OpenClaw Skill
Name: opencli-skills
Version: 1.0.1
The skill bundle provides a wrapper for 'opencli', a tool designed to automate data extraction and interaction across over 30 platforms (including WeChat, Twitter, Discord, and Boss Zhipin) by leveraging a 'Browser Bridge' extension to reuse active Chrome login sessions. While the functionality aligns with the stated purpose, the skill grants the agent broad, high-risk capabilities such as reading private messages, sending posts, and accessing sensitive financial or recruitment data without API restrictions. This extensive access to authenticated user sessions across multiple services represents a significant risk for data exfiltration or unauthorized account activity if the agent is misdirected. (Files: SKILL.md, _meta.json; Dependency: @jackwener/opencli)
Capability Assessment
Purpose & Capability
The name/description (fetch website data by reusing Chrome login sessions) aligns with the SKILL.md: it instructs using @jackwener/opencli + Browser Bridge and lists many site-specific commands. However the command set includes active actions (send messages, control desktop chat apps, download video) in addition to read-only fetches; that breadth is larger than the simple 'fetch data' description suggests.
Instruction Scope
Runtime instructions instruct dumping DOM, reading chat messages, sending messages, and interacting with desktop/web app UIs via the Browser Bridge — operations that can access and exfiltrate authenticated/private content. The skill expects reuse of Chrome sessions (i.e., access to cookies/session state) which is functionally necessary for its purpose but also grants broad access to any logged-in sites. The SKILL.md does not place clear limits or safeguards on these powerful operations.
Install Mechanism
This is an instruction-only skill (no install spec in the registry). It tells users to run `npm install -g @jackwener/opencli` and to install the Browser Bridge Chrome extension. Installing a global npm package and a browser extension is a typical path for this functionality but carries risk (supply-chain/browser extension permissions). The SKILL.md also references additional runtime tooling (e.g., yt-dlp for downloads) that is not formalized in the metadata.
Credentials
The skill requests no environment variables or credentials, but its operation depends on accessing your Chrome login sessions via a browser extension — effectively granting access to your authenticated accounts. That access is proportionate to scraping account-scoped data, but the SKILL.md does not enumerate or limit which sites, nor does it require explicit per-site consent in the instructions. Also, some commands mention external binaries (yt-dlp) which are not declared in the top-level requirements.
Persistence & Privilege
The skill is not always-enabled and does not request elevated platform-level persistence; it is instruction-only and would operate only when invoked. It also does not declare modifications to other skills or system-wide config.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install opencli-skills - After installation, invoke the skill by name or use
/opencli-skills - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
No changes detected in this version.
- Version updated from 1.0.0 to 1.0.1.
- No modifications were made to any files.
- Functionality and documentation remain unchanged.
v1.0.0
opencli 1.0.0
- Initial release of the OpenCLI skill.
- Fetch data from popular sites like Twitter, Reddit, Bilibili, Zhihu, Xiaohongshu, YouTube, and more, without needing API keys.
- Leverages Chrome with Browser Bridge extension to reuse login sessions.
- Supports dozens of built-in commands across many platforms, with customizable output formats and filters.
- See detailed command and parameter lists for each supported platform.
Metadata
Frequently Asked Questions
What is opencli?
Fetch data from social, video, and news websites using OpenCLI by reusing Chrome login sessions without API keys. It is an AI Agent Skill for Claude Code / OpenClaw, with 211 downloads so far.
How do I install opencli?
Run "/install opencli-skills" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is opencli free?
Yes, opencli is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does opencli support?
opencli is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created opencli?
It is built and maintained by 张贝 (@hellotombruce); the current version is v1.0.1.
More Skills