ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

图图智控（TUTU Smart Control）

v1.3.3

图图智控（TUTU Smart Control）— 远程控制 Android 手机，执行 GUI 自动化、系统管理和日常任务。当用户提到以下任何内容时使用此 skill： - 控制手机、操作手机、手机上做某事 - 打开手机上的 App（微信、抖音、支付宝、淘宝等） - 手机截图、查看手机屏幕 - 在手机上点击、输...

⭐ 0· 547·1 current·1 all-time
by@zhaojiaqi
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
Name/description (remote Android control) align with requirements: a single API token (TUTU_API_TOKEN) and HTTP calls to the TUTU API. No unrelated environment variables, binaries, or install steps are requested. The tool list (screenshot, click, read_sms, etc.) matches the stated remote-control use case.
ℹ
Instruction Scope
SKILL.md instructs the agent to make POST requests to https://www.szs.chat/api/phone_action.php using the injected TUTU_API_TOKEN and to fetch/analyze screenshot URLs returned by the API. That is coherent with remote-control functionality. Two points to note: (1) many operations are sensitive (read SMS/contacts/location, send SMS, uninstall, delete files) but the skill marks them as requiring platform-level confirmation — this relies on the platform enforcing confirmation. (2) The guidance to 'analyze the screenshot and decide next steps' grants the agent broad discretion; that is expected for an automation skill but increases the importance of platform confirmation controls and user oversight.
✓
Install Mechanism
Instruction-only skill with no install spec and no code files to execute. Lowest-risk installation model: nothing is downloaded or written to disk by the skill itself.
✓
Credentials
Only one environment variable (TUTU_API_TOKEN) is required and it's directly justified by the API-based remote-control purpose. No unrelated secrets or system config paths are requested.
✓
Persistence & Privilege
always:false (default) and the skill does not request persistent system-level changes or access to other skills' configs. It can be invoked autonomously by the agent (platform default) — that is normal but means the agent could attempt actions unless platform confirmation gates are applied for sensitive operations.
Assessment
This skill appears internally consistent: it uses a single injected API token to call a named API and exposes expected phone-control actions (screenshot, click, read SMS, send SMS, location, file ops on /sdcard, etc.). Before installing: (1) only set TUTU_API_TOKEN if you trust the provider/domain (https://www.szs.chat) and the hardware device, as the token grants remote control of the device; (2) confirm your platform enforces the declared 'confirmation' prompts for sensitive actions (send SMS, read contacts/sms, uninstall/delete, GPS mock, etc.); (3) treat screenshots and returned data as potentially sensitive — the skill directs the agent to fetch screenshot URLs from the provider; ensure those endpoints are trusted and that the platform won’t leak the token in logs; (4) consider rotating/revoking the token if you stop using the device. If you need higher assurance, ask the publisher for an authoritative homepage/legal info and independent verification of the TUTU service and confirmation enforcement.

Like a lobster shell, security has layers — review code before you run it.

latestvk97dyn3qk1874fk492k1kxqwdd81wr6b

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

📱 Clawdis
EnvTUTU_API_TOKEN
Primary envTUTU_API_TOKEN

Comments