ClawLock-Rank

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed leaderboard uploader for ClawLock scan scores, with user confirmation and limited uploaded fields.

Install only if you are comfortable sending a sanitized ClawLock score submission to the configured ClawLockRank service. Review the preview before approving upload, choose a nickname you are comfortable making public, and avoid uploading from environments where the device fingerprint or finding titles could reveal sensitive operational details.

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)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 95% confidence
Finding: The skill instructs the agent to run local Python scripts, invoke shell commands, read scan output, and perform network uploads, but the manifest declares no permissions. This creates a trust and review gap: an agent or platform may authorize the skill without surfacing that it can access local data, environment context, and exfiltrate scan-derived information over the network.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal