ClawHub
Back to skill

Security audit

Synomega Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local chemistry tool with expected model downloads and configuration knobs, but users should treat route planning for hazardous compounds carefully.

Install only if you are comfortable with a third-party Python package and a first-use download of several hundred MB from remote mirrors. In air-gapped, privacy-sensitive, or reproducibility-critical environments, pre-fetch assets or point the skill at vetted local model and stock files. Apply your host safety policy before using it for toxic, controlled, explosive, or otherwise high-risk compounds.

SkillSpector

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
87% confidence
Finding
The skill explicitly documents reliance on environment variables such as SYNOMEGA_CACHE, SYNOMEGA_MIRROR, SYNOMEGA_MODEL, SYNOMEGA_STOCK, and related toggles, but the metadata does not declare corresponding permissions/capabilities. Undeclared env access weakens transparency and policy enforcement because execution behavior can be altered by external environment configuration, including redirecting model or stock paths and controlling first-run network-fetch behavior.

Static analysis

No suspicious patterns detected.