Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
yuketang
v1.0.327853提供雨课堂账户和班级相关查询服务，包括用户ID、开班列表、班级数据、预警名单、今日授课及作业公告完成情况查询等。
⭐ 1· 86·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The skill's declared purpose (rain-classroom account and class queries) legitimately requires a YUKETANG_SECRET, and both SKILL.md and package.json expect that env var. However the registry metadata in the provided summary incorrectly lists 'Required env vars: none' and 'Primary credential: none' — that's an inconsistency that could mislead users about required credentials.
Instruction Scope
SKILL.md instructs the agent/user to obtain YUKETANG_SECRET, set it as an environment variable and run the bundled setup.sh or setup.js. The scripts register an MCP endpoint via npx mcporter and, if available, call the MCP to report install telemetry. The instructions also say the agent should help update env vars for users (i.e., modify shell/project config) which grants the skill practical ability to persist credentials on disk — this is within the skill's purpose but worth explicit user consent.
Install Mechanism
There is no formal install spec, but setup.sh and setup.js use 'npx mcporter@0.8.1' to add a project-scoped MCP config. Using npx pulls a package from npm at runtime (moderate risk but common). The MCP URL is a named domain (open-ai.xuetangx.com), not an IP or shortener — no obvious shady download host. setup scripts also print a manual config fallback. Overall expected, but npm invocation and writing config files elevates risk slightly.
Credentials
The only credential used is YUKETANG_SECRET, which is appropriate for this integration. However the skill will (via mcporter config) store the authorization header (Bearer <secret>) in a local MCP config (project-scoped .mcp/config.json or similar), persisting the secret on disk in plaintext. The metadata omission of the required env var is also problematic because users may not realize a secret is required or stored.
Persistence & Privilege
The skill does not set always:true and does not modify other skills, but the setup scripts add a project-scoped MCP server configuration (persisting credentials) and perform a silent install telemetry call (npx mcporter call yuketang-mcp claw_report). Persisting the secret to project config and sending telemetry to the MCP endpoint are normal for an MCP integration but constitute persistent presence and outbound network activity that users should accept knowingly.
What to consider before installing
This skill appears to be a legitimate connector for 雨课堂 (Yuketang) but there are a few things to check before installing: 1) Metadata mismatch — the repository summary incorrectly listed no required env vars, but the skill requires YUKETANG_SECRET. Treat that as a red flag and make sure you understand why the metadata is wrong. 2) The setup scripts will persist your secret into the local MCP client config (project-scoped .mcp/config.json or similar) and will register an MCP server (open-ai.xuetangx.com) using that secret. If you run the installer, inspect the created config file and ensure you are comfortable storing the secret in the project directory (anyone with filesystem access may read it). 3) setup scripts use 'npx mcporter' (runtime npm package fetch) and perform a silent telemetry 'install' call to the remote MCP server; review what mcporter does and whether you trust the remote domain. Recommended actions: a) Do not run the scripts blindly. Prefer the manual configuration path SKILL.md prints and add the MCP entry yourself after inspecting its content. b) Inspect the output of 'npx mcporter config add ...' and the resulting config file; if you prefer, store the secret in a secure system-level credential store instead of a project file. c) Confirm you trust the MCP host (https://open-ai.xuetangx.com) and the secret source URL (https://www.yuketang.cn/ai-workspace/open-claw-skill). d) If you install and later decide to remove the skill, remember to remove the MCP config and rotate the YUKETANG_SECRET to limit exposure. e) If you want to proceed but are cautious, run the setup scripts in an isolated/sandboxed project directory and monitor network calls. If you need more evidence (e.g., confirmation of what is persisted in your specific client paths), provide your client type (OpenClaw/CodeBuddy/Cursor) and I can point to the exact config file to inspect.setup.js:35
Shell command execution detected (child_process).
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Like a lobster shell, security has layers — review code before you run it.
latestvk973tn1mm1a0x1fdvt0ef3g6v5844sfc
License
MIT-0
Free to use, modify, and redistribute. No attribution required.