中文工具包

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: chinese-toolkit Version: 1.0.1 The skill is suspicious due to a significant supply chain risk identified in `install.sh`, which clones the project from `https://github.com/utopia013-droid/luxyoo.git` instead of the declared `https://github.com/openclaw/chinese-toolkit` (or `openclaw-cn/chinese-toolkit` as seen in other documentation). This inconsistency in the source repository is a critical vulnerability. Additionally, `openclaw_integration.py` exposes `_handle_ocr` and `_handle_convert` methods that likely wrap command-line tools (like `tesseract` and `opencc` mentioned in `SKILL.md` and `scripts/install_deps.sh`). Without the full implementation of these methods, there is a high risk of shell injection if input parameters are not properly sanitized, leading to a potential Remote Code Execution vulnerability.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

ConcernMedium Confidence
ASI02: Tool Misuse and Exploitation
What this means

If an agent follows these bundled release instructions outside a maintainer workflow, it could publish or update skills using the user's account.

Why it was flagged

A Chinese text-processing skill bundles instructions that encourage immediate execution of a publishing workflow. Publishing to ClawHub is a high-impact account action and is not needed for ordinary Chinese text/OCR/translation use.

Skill content
## 🚀 立即执行:一键完成所有发布 ... .\complete_release.ps1 ... npx clawhub publish . --version $version
Recommendation

Do not let the agent run the release or publishing guides unless you are intentionally maintaining this package. The publisher should remove these release guides from the distributed runtime skill or clearly mark them as maintainer-only.

ConcernMedium Confidence
ASI03: Identity and Privilege Abuse
What this means

Following these instructions could grant or use GitHub/ClawHub account authority for publishing, which goes beyond processing text or translation.

Why it was flagged

The bundled documentation asks for GitHub/ClawHub authentication and describes account privileges unrelated to the toolkit's stated Chinese-language processing purpose.

Skill content
npx clawhub login --token YOUR_GITHUB_TOKEN ... GitHub OAuth ... 可以读取用户信息和邮箱 ... 可以发布技能到市场
Recommendation

Only authenticate GitHub/ClawHub for this package if you are deliberately publishing it. End users installing the toolkit should not provide GitHub tokens or OAuth access for normal use.

ConcernMedium Confidence
ASI08: Cascading Failures
What this means

If executed unintentionally, these commands could disrupt a GitHub repository or overwrite shared history.

Why it was flagged

The emergency troubleshooting section documents force-push commands. In the wrong context, these can overwrite repository history and propagate changes to shared projects.

Skill content
git push github master --force ... git push github v1.0.0 --force
Recommendation

Treat force-push instructions as maintainer-only and require explicit human confirmation before any repository-mutating command.

What this means

Text submitted for translation may be sent to third-party translation APIs.

Why it was flagged

The translation feature sends user-provided text to an external translation provider. This is expected for a translation skill, but it is a sensitive data flow users should notice.

Skill content
url = "https://fanyi-api.baidu.com/api/trans/vip/translate" ... 'q': text
Recommendation

Avoid translating confidential documents through external providers unless you are comfortable with that provider's data handling. Prefer local translation modes for sensitive content if available.

What this means

Users may need to provide third-party API keys for some features, and those keys should be protected.

Why it was flagged

The skill documents provider API credentials for translation/cloud services. These credentials are purpose-aligned, but the registry metadata declares no required environment variables or primary credential.

Skill content
export BAIDU_TRANSLATE_APP_ID="your_app_id" ... export TENCENT_CLOUD_SECRET_KEY="your_secret_key"
Recommendation

Store provider credentials securely, scope them to the minimum needed permissions, and do not share them with unrelated workflows.

What this means

Some processed results could persist temporarily in a cache or logs.

Why it was flagged

The skill configuration enables caching. Caching is reasonable for performance, but for translation/OCR/document workflows it may retain derived content unless users understand where and how it is stored.

Skill content
"cache": { "enabled": true, "ttl": 3600, "max_size": 1000, "cleanup_interval": 300 }
Recommendation

Review cache and logging settings before processing sensitive documents, and disable or clear caches when privacy matters.

What this means

Installing dependencies can add software to the user's machine and may require elevated privileges for system packages.

Why it was flagged

The skill asks users to install Python and system dependencies. This is expected for OCR/audio/text tooling, but it changes the local environment and should be user-approved.

Skill content
pip install -r requirements.txt ... sudo apt-get install tesseract-ocr tesseract-ocr-chi-sim ffmpeg ... brew install tesseract tesseract-lang ffmpeg
Recommendation

Run installation commands only after reviewing requirements and prefer a virtual environment for Python packages.