Security audit

Go安全质量检查专业版

Security checks for vulnerabilities and agentic risk

Overview

This Go security skill is mostly purpose-aligned, but it includes repository-changing remediation commands without clear opt-in or containment guidance.

Install only if you want an agent to run Go security tools over approved repositories. Treat scans and SARIF reports as potentially sensitive, and require explicit confirmation before dependency updates, batch scans outside the current repo, tool installation, or uploads to GitHub Code Scanning.

SkillSpector

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (4)

Intent-Code Divergence

Medium

Confidence: 91% confidence
Finding: The skill's security note claims user input should not be concatenated into shell commands, but earlier examples interpolate path-like values such as project directories and derived names directly into command execution flow. In an agent setting with exec enabled, this contradiction can normalize unsafe implementation patterns and lead to command injection or unsafe filesystem targeting if future invocations substitute less-trusted input.

Intent-Code Divergence

Low

Confidence: 72% confidence
Finding: The skill states outputs will not contain sensitive data, yet it documents credential-bearing workflows and external SARIF uploads without corresponding controls for redaction, logging suppression, or secret-safe handling. This creates a misleading assurance that may cause operators to expose repository metadata, findings, or tokens in logs and outbound integrations.

Vague Triggers

Medium

Confidence: 83% confidence
Finding: The activation wording is broad enough to trigger this skill for generic 'security detection,' 'compliance audit,' 'quality inspection,' and 'encryption protection' tasks, while the skill has exec capability and can perform impactful actions. Ambiguous triggering increases the chance an agent invokes scanning or modification workflows in the wrong context, against unintended directories, or without sufficiently specific user approval.

Missing User Warnings

High

Confidence: 95% confidence
Finding: The skill includes an automatic remediation workflow that runs `go get` and `go mod tidy`, which changes dependency state and project files, but it does not present a prominent warning or explicit confirmation requirement before modification. In agent-driven environments, this can lead to unintended codebase changes, dependency drift, broken builds, or unauthorized edits across repositories.

Static analysis

No suspicious patterns detected.