ClawHub
Back to skill

Security audit

golang-security

Security checks across malware telemetry and agentic risk

Overview

This Go security skill is coherent and purpose-aligned, with a couple of imperfect security examples users should review before copying.

Reasonable to install for Go security work. Treat its examples as guidance rather than drop-in code, especially the logging sanitizer and unsafe overlap-check example, and approve audit/fix modes only when you want the agent to inspect and modify your Go code.

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)

Intent-Code Divergence

Medium
Confidence
98% confidence
Finding
This is a true issue in the guidance: the example claims to prevent log injection, but the sanitizer explicitly preserves newline and tab characters. Newlines are a primary vehicle for log forging and record-splitting attacks, so readers who copy this code may still allow attackers to inject fake log entries or corrupt downstream parsing and monitoring.

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
This guidance is internally inconsistent: it states that `unsafe` must not be used in application code, then later presents a `Good` example that relies on `unsafe.Pointer` for overlap detection. In a security best-practices skill, contradictory examples can normalize prohibited memory-unsafe techniques and lead developers to copy unsafe pointer arithmetic into production code, undermining the stated safety rule.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
references/secrets.md:20