Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Online Course Creator
v1.0.0AI驱动的一键生成完整在线课程大纲、视频脚本、测验及营销材料，支持快速高效课程创作与推广。
⭐ 1· 201·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description match the included code: index.js and example.js implement course outline, video script, quiz and marketing material generation. No unexpected binaries, services, or credentials are requested.
Instruction Scope
SKILL.md usage instructions are scoped to install and invoke the skill and include trigger phrases; however a pre-scan detected unicode-control-chars inside SKILL.md (prompt-injection pattern). The instructions do not ask for unrelated files/credentials, but hidden control characters can be used to manipulate downstream prompt processing — review raw SKILL.md bytes.
Install Mechanism
Registry lists no install spec (instruction-only), but the package includes package.json, index.js and example.js for Node.js usage. There are no external download URLs or extract steps in the files provided. This is low technical risk, but the mismatch (no install spec vs. included code) is worth being aware of.
Credentials
The skill requests no environment variables, no credentials, and code only uses local filesystem (fs, path). There are no obvious requests for unrelated secrets or external APIs in the provided files.
Persistence & Privilege
Flags show always:false and user-invocable:true. The skill does not request persistent elevated privileges or attempt to modify other skills or system-wide agent configuration in the provided code.
Scan Findings in Context
[unicode-control-chars] unexpected: Hidden/unicode control characters were detected in SKILL.md. These are not necessary for the described course-generation functionality and are commonly used in prompt-injection attempts to alter how text is parsed or executed. The rest of the code appears local-only (no network calls), but you should inspect the raw SKILL.md (hex/byte view) and remove or validate any control characters before using.
What to consider before installing
What to do before installing:
- Inspect SKILL.md in a raw/hex editor for hidden unicode control characters and remove them. The scanner flagged this as a prompt-injection signal.
- Review index.js and example.js locally (they appear to only use fs/path and generate content) and run them in an isolated sandbox or container first.
- Verify the package origin (the skill lists a GitHub URL and OpenClaw Team author; confirm the repository and commits match the published package). Do not run 'clawhub install' or any install commands on a production machine until you've validated locally.
- Run a quick static check for network calls (search for http, https, fetch, axios, net, child_process.exec) — the provided files show none, but confirm no hidden dynamic requires are present.
- If you accept the risk, run the code in a disposable environment (container/VM) and inspect outbound network traffic during execution. If you are not comfortable with these steps, do not install.Like a lobster shell, security has layers — review code before you run it.
latestvk97fhmc71mq0dqnwa4ewjs0fyh836d36
License
MIT-0
Free to use, modify, and redistribute. No attribution required.