Back to skill

Security audit

电商图裂变 | 图片裂变 | 多版本生成 | 广告素材裂变 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for generating marketing image variations, but it tells the agent to bootstrap and upgrade unpinned CLI and Node tooling on the host without clear user confirmation.

Install only if you are comfortable letting the agent install or upgrade qhkit/Node tooling and configure a Qinghu token for the CN service. Prefer preinstalling a reviewed, pinned qhkit version yourself and asking the agent not to run global npm, npx, or Node bootstrap commands automatically.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
Findings (7)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description says this skill 'must trigger' for broad classes of user requests, creating an aggressive auto-invocation rule. In context, that can cause the agent to enter a workflow that installs software, requests tokens, or contacts external services even when the user only asked generally about generating variants.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill goes far beyond image-generation guidance by instructing the agent to install Node.js, globally install a CLI, and later upgrade it. This materially expands the attack surface and permits persistent modification of the host environment, including fetching and executing remote code from package registries and download mirrors.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The skill recommends falling back to npx @iqinghu/qhkit <命令> without pinning a version, which causes execution of whatever package version is current at invocation time. That introduces a supply-chain risk: behavior can change unexpectedly or a compromised upstream release could execute arbitrary code in the agent environment.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
94% confidence
Finding

This duplicate finding points to the same persistence behavior: installing Node into a user directory and changing PATH for subsequent shells. Even with checksum verification, it still establishes persistent execution capability beyond the intended image-variation task and widens the blast radius of any upstream compromise or operator mistake.

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

Linux 二进制安装(装到用户目录,无需 root;先校验官方 SHA256 再解包):

cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | sha256sum -c - mkdir -p "$HOME/.local/lib" && tar -xJf /tmp/node-v22.22.3-linux-x64.tar.xz -C "$HOME/.local/lib" export PATH="$HOME/.local/lib/node-v22.22.3-linux-x64/bin:$PATH"

text

Session Persistence

Medium
Category
Rogue Agent
Confidence
94% confidence
Finding

This duplicate finding points to the same persistence behavior: installing Node into a user directory and changing PATH for subsequent shells. Even with checksum verification, it still establishes persistent execution capability beyond the intended image-variation task and widens the blast radius of any upstream compromise or operator mistake.

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

Linux 二进制安装(装到用户目录,无需 root;先校验官方 SHA256 再解包):

cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | sha256sum -c - mkdir -p "$HOME/.local/lib" && tar -xJf /tmp/node-v22.22.3-linux-x64.tar.xz -C "$HOME/.local/lib" export PATH="$HOME/.local/lib/node-v22.22.3-linux-x64/bin:$PATH"

text

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The instructions require the domestic CN site configuration and say to preserve intl only if the user explicitly says they use the international account. This imposes a locale/site choice by default rather than offering the user a language/region option, which conflicts with the requirement not to force a specific locale without user opt-in.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The '何时触发' section provides only a couple of positive examples, including a very general request like '多出几版不同风格的', but does not specify when the skill should not be used. Without exclusions, similar everyday requests could accidentally match this skill instead of a more appropriate image-editing or design tool.

Content

No source excerpt is available for this finding.