Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 70% confidence
- Finding
- Without declared permissions the skill's intent is opaque and cannot be validated.
Security audit
Take Back My Latte From Claude
Security checks across malware telemetry and agentic risk
Overview
This skill locally analyzes user-provided Claude billing exports and does not show hidden upload, credential, persistence, or privilege behavior.
Installers should understand that this skill will read Claude billing export JSON files they provide and summarize cost data locally. The skill also points to an external website for follow-up action; review that site separately before entering or uploading any sensitive billing data there.
SkillSpector
By NVIDIA
Vulnerability Patterns
- MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)
VirusTotal
65/65 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.