Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Ai File Organizer
v3.0.0AI 智能文件整理 - 批量重命名、自动分类、智能归档（异步引擎 + 云同步）
⭐ 0· 158·0 current·0 all-time
by@sxliuyu
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's name/description (file organizer with AI analysis and cloud sync) aligns with the included code and docs. However the published registry 'Requirements' block said no required environment variables, while _meta.json and documentation list ALIYUN_ACCESS_KEY, ALIYUN_SECRET_KEY and BAIDU_ACCESS_TOKEN and network endpoints — an inconsistency in declared requirements. Requesting cloud provider credentials is coherent with cloud-sync features, but the mismatch between declared/required env vars is unexplained and worth questioning.
Instruction Scope
SKILL.md and the code instruct the agent/user to scan user directories, compute file hashes, persist caches under the user's home (~/.ai-organizer), and (optionally) send file contents/metadata to external services for AI analysis and upload to cloud providers. Sending file contents to an external LLM/service (dashscope) and to cloud endpoints is functionally expected for 'AI content analysis' and 'cloud sync', but it is a high privacy surface: these instructions permit reading arbitrary user files in user directories and transmitting their contents to external endpoints if AI or cloud sync is enabled. The documentation also claims 'sensitive path exclusion' and 'encrypted storage' — you should verify those are actually enforced in the code.
Install Mechanism
There is no formal install spec in the registry (instruction-only install). The Quickstart suggests pip installing dependencies (aiofiles, aiomultiprocess, tqdm, pyyaml and optionally dashscope). This is standard for a Python tool and less risky than arbitrary binary downloads. Still: installing third-party packages (especially less-known ones like 'dashscope') carries the usual supply-chain risk — inspect those packages before installing.
Credentials
The feature set justifies cloud credentials (ALIYUN_*, BAIDU_* and OneDrive equivalents) and possibly an API key for an LLM provider; those appear in README/_meta.json. However the top-level registry summary claimed 'Required env vars: none' while _meta.json lists environmentVariables. Also network access and endpoints (dashscope.aliyuncs.com, api.aliyundrive.com, pan.baidu.com) are declared in metadata: granting credentials and network access would allow uploading user files and sending content to external LLMs. Require only the credentials you intend to use, and avoid entering secrets unless you trust the code and host.
Persistence & Privilege
The skill persists state to the user's home (~/.ai-organizer) (cache file, logs), writes to user-specified target folders and creates version folders. always: false (not force-enabled). Persistence is expected for this tool, but review and control where data is stored, and ensure the 'sensitive path exclusion' logic works before running broadly or with cloud sync enabled.
What to consider before installing
This skill appears to implement the advertised file-organization features, including optional AI analysis and cloud-sync which will read user files and may transmit their contents to external services. Before installing or providing any credentials: 1) Verify the skill source/trustworthiness (the package homepage/source is missing). 2) Review scripts/organizer.py (particularly network calls and any code that uploads or posts file content) or run it in a sandboxed account. 3) If you need only local features, keep ai_analysis and cloud_sync disabled and do not set cloud/LLM credentials. 4) If you enable cloud sync or AI analysis, create and scope credentials with least privilege and prefer temporary/test accounts. 5) Confirm that 'sensitive path exclusion' and config encryption are actually implemented (search code for exclusion patterns and encryption calls). 6) Inspect and vet third-party dependencies (e.g., 'dashscope') before pip installing. If you want, I can scan organizer.py for specific network calls and the exact places file contents are transmitted and summarize those lines for you.Like a lobster shell, security has layers — review code before you run it.
aivk97ez14470werc19sgah6bhw3983131cautomationvk97ez14470werc19sgah6bhw3983131cfilevk97ez14470werc19sgah6bhw3983131clatestvk977xjvc6f2qej567r6jf1mdrx831hkh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.