Scan Report
25 /100
emwstudio
电磁波Studio ComfyUI workflow integration skill for RunningHub platform - supports video generation and audio synthesis workflows
A legitimate ComfyUI workflow integration skill for RunningHub platform with minor documentation quality issue (hardcoded IP in example response). No malicious behavior detected in actual code.
Safe to install
Consider removing hardcoded IP address from SKILL.md documentation example and replacing with placeholder. Otherwise, the skill is safe for use.
Findings 1 items
| Severity | Finding | Location |
|---|---|---|
| Low | Hardcoded IP address in documentation example Doc Mismatch | SKILL.md:130 |
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Filesystem | READ | READ | ✓ Aligned | get_env.py:29 - reads ~/.openclaw/openclaw.json |
| Network | READ | READ | ✓ Aligned | All scripts only make HTTPS requests to www.runninghub.cn or www.runninghub.ai |
| Shell | WRITE | WRITE | ✓ Aligned | SKILL.md declares use of OpenClaw exec tool for running Python scripts |
1 High 3 findings
High IP Address 硬编码 IP 地址
222.186.161.123 SKILL.md:130 Medium External URL 外部 URL
https://www.runninghub.cn/?inviteCode=6bfdf1c0 SKILL.md:232 Medium External URL 外部 URL
https://www.runninghub.ai/?inviteCode=6bfdf1c0 SKILL.md:232 File Tree
10 files · 27.1 KB · 930 lines Python 8f · 682L
Markdown 1f · 235L
JSON 1f · 13L
├─
▾
data
│ └─
workflows.json
JSON
├─
▾
scripts
│ ├─
config_api_key.py
Python
│ ├─
config_host.py
Python
│ ├─
create_task.py
Python
│ ├─
get_account_info.py
Python
│ ├─
get_env.py
Python
│ ├─
get_workflow_info.py
Python
│ ├─
poll_task.py
Python
│ └─
query_task.py
Python
└─
SKILL.md
Markdown
Security Positives
✓ All network requests go to legitimate RunningHub endpoints (www.runninghub.cn or www.runninghub.ai)
✓ No obfuscation techniques detected - all Python code is plaintext
✓ No credential exfiltration - API keys are stored locally in OpenClaw config and used only for RunningHub API calls
✓ No subprocess, eval, exec, compile, or __import__ calls in actual scripts
✓ No access to sensitive paths like ~/.ssh, ~/.aws, or .env files
✓ Shell execution is properly declared in SKILL.md documentation
✓ No base64 encoding/decoding or anti-analysis techniques found
✓ API keys are not logged or transmitted to third parties