Security audit

测试复盘

Security checks for vulnerabilities and agentic risk

Overview

This skill is a QA retrospective template that uses limited read-only workspace access and does not show hidden persistence, exfiltration, or unsafe actions.

Install this only if you want an agent to help review QA, defect, and iteration data in your workspace. Use explicit prompts such as testing retrospective or missed-defect analysis, and review any suggested process changes with the team before adopting them.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The skill’s trigger conditions are broad enough to match ordinary requests like summaries, retrospectives, or lessons learned, which can cause unintended activation. In an agent system, over-broad activation can route users into a rigid workflow they did not request, increasing the chance of inappropriate file reads, unnecessary process guidance, or context confusion.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The example trigger phrase “这个问题又发生了” is highly generic and can appear in many unrelated contexts, making accidental skill activation likely. This is risky because it may steer the agent into retrospective/root-cause behavior without clear user consent or sufficient QA context.

Static analysis

No suspicious patterns detected.