Back to skill

Security audit

test-publish

Security checks for vulnerabilities and agentic risk

Overview

The skill appears intended to automate product distribution, but it embeds login credentials, uses insecure HTTP, and can submit changes without reliable validation or confirmation.

Review this skill before installing. It may trigger commerce-related distribution actions using a shared embedded account over an unencrypted connection, and it lacks safeguards to confirm the final category, store, and price before sending. Do not use it with real stores until credentials are removed, HTTPS is used, the script path is fixed, and submission is gated by validated selections and explicit user approval.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/auto_distribution.py:50
Finding

Hardcoded Credentials Transmitted over Plaintext HTTP

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/auto_distribution.py:32
Finding

Fail-Open Submission after Required Selection Failures

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
SKILL.md:25
Finding

Workflow References a Nonexistent Script Path

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The finding indicates a material mismatch between the declared skill purpose and the underlying behavior: the skill claims to publish products to Ozon, but the referenced implementation reportedly logs into an external site with hardcoded credentials, performs browser automation on a fixed domain, and stores local screenshots without clearly implementing the advertised listing workflow. This is dangerous because users and downstream systems may grant the skill trust and permissions based on a benign description while it performs broader or different actions, creating risk of credential misuse, unauthorized site interaction, and unreviewed data capture.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The manifest and operational instructions are entirely written in Chinese and do not indicate that users may interact in other languages or choose a locale. Under the policy, language/locale restrictions should be opt-in or clearly justified as region-specific; this file does not state such a justification.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

清单描述声称该技能会把商品从“跨睿优质货盘”自动铺货到 Ozon 平台,但本文件中的实际行为只是访问一个固定内网/裸 IP 页面、使用硬编码账号登录、选择品类/店铺/价格区间、点击“发送”并截图。代码中没有出现 Ozon 平台交互、商品数据读取/映射、发布结果校验等与“铺货到 Ozon”直接对应的实现迹象,因此描述与实际行为存在明显语义落差。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
99% confidence
Finding

The script submits hardcoded credentials over plain HTTP to a fixed IP address, exposing authentication data to interception and reuse by anyone on the network path. Hardcoded secrets also encourage credential reuse, make rotation difficult, and risk leakage through source control or logs.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The script unconditionally clicks a generic 'send' button after login and selection without validating the target action or requiring user confirmation. In an automation skill that claims to distribute products, this could trigger unintended publication, submission, or other state-changing actions at scale if selectors drift or parameters are wrong.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

顶部文档特别强调“新策略:使用JavaScript直接操作Vue组件”,这会让读者理解为代码会注入或执行 JavaScript 来直接操纵 Vue 组件状态。实际实现中未见 evaluate、脚本注入或组件实例访问,只有常规的 page.fill 和 page.click,因此文档对实现方式构成主动误导。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The module docstring and all user-facing console messages are exclusively in Chinese, which imposes a language choice on users without opt-in or documented justification. The policy calls for flagging language or locale constraints when the skill forces a specific language without offering a choice.

Content

No source excerpt is available for this finding.