Trusted — Risk Score 5/100
Last scan:1 day ago Rescan
5 /100
replyher
AI communication coach for dating, workplace, and social messaging with multi-language support (Chinese/English)
ReplyHer is a pure text-based communication coach with no code execution, credential access, or data exfiltration. The only script present (publish.sh) is a legitimate CI/CD deployment tool for developers, not the AI agent itself.
Skill Namereplyher
Duration27.3s
Enginepi
Safe to install
This skill is safe to use. The publish.sh script requires shell access but is a developer tool for deployment only, not accessible to the AI agent.
ResourceDeclaredInferredStatusEvidence
Filesystem NONE NONE SKILL.md contains no filesystem access declarations
Network NONE NONE No network calls in AI behavior
Shell NONE NONE publish.sh is developer tooling, not AI-accessible
Environment NONE NONE No environment variable access declared or observed
Skill Invoke NONE NONE No cross-skill invocation observed
Clipboard NONE NONE No clipboard operations
Browser NONE NONE No browser automation
Database NONE NONE No database access
2 findings
🔗
Medium External URL 外部 URL
https://replyher.com
README.md:45
🔗
Medium External URL 外部 URL
https://clawhub.ai/skills/replyher
publish.sh:26

File Tree

3 files · 18.2 KB · 511 lines
Markdown 2f · 485L Shell 1f · 26L
├─ 🔧 publish.sh Shell 26L · 782 B
├─ 📝 README.md Markdown 89L · 3.1 KB
└─ 📝 SKILL.md Markdown 396L · 14.4 KB

Security Positives

✓ Pure text-based AI skill with no code execution capabilities
✓ No credential harvesting or sensitive data access
✓ No network requests or data exfiltration
✓ No obfuscation techniques (base64, eval, etc.)
✓ Clear documentation matching actual behavior
✓ Safety guidelines present for respecting boundaries
✓ publish.sh is legitimate CI/CD tooling, not malicious