Security audit

Clawnalyst

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a legitimate analytics/API integration, but it can change remote account profile and payout settings and has disclosure mismatches that warrant review before installation.

Install only if you intend to let the skill contact both the API and MCP service and potentially change Clawnalyst account settings. Before use, verify the exact endpoints, require explicit confirmation for any profile or payout-wallet update, and prefer a version that fixes the shell quoting issue in the helper scripts.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (2)

Description-Behavior Mismatch

Medium

Confidence: 92% confidence
Finding: The skill metadata describes posting signals and reading stats, but the body also documents modifying the user's remote profile, including payout wallet and active status. Undisclosed state-changing functionality increases the chance of unauthorized or unexpected account changes because reviewers may not realize the skill can alter account configuration.

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The security manifest claims only api.clawnalyst.com is contacted, but the documentation also directs direct use of the MCP server at mcp.clawnalyst.com. This mismatch undermines trust in the manifest and can bypass endpoint allowlisting or review assumptions about where credentials and user data may be sent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.