Back to skill

Security audit

产品需求证据调研

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious because it repeatedly runs a changeable CLI package without pinning its version; the available scan is partial.

Consider the unpinned CLI execution risk before installing, especially because the login flow supplies an API key to that package. Prefer a reviewed, fixed package version with verified integrity. The scan is marked partial: one local reference could not be resolved, so the evidence is incomplete.

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
  • 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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (13)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 27)May include surrounding context.

md
` 与来源 `environment` / `file` / `null`,不输出 Key。安装或升级时,已有凭证可读、没有已知鉴权失败且用户未明确要求更换 Key,就复用原来源继续原任务,不重新登录、不迁移或改写已有凭证;命令、权限或文件格式错误按实际错误处理,不一律判为没配置。已认证的 hosted MCP 沿用当前连接鉴权,不要求另配本地 CLI。
- 仅首次配置、凭证缺失或失效,或用户明确要求更换 Key 时才录入:用户已在安装提示词或当前对话中提供 Key 并授权配置时,用 `npx -y socialdatax-skills@latest auth login --stdin` 接收这枚 Key 的标准输入,由 CLI 长期保存;不要求用户重复手动录入,不把 Key 作为 CLI 参数或写入临时脚本文件。没有提供 Key 时,引导用户从上述官网获取,在本地终端运行 `npx -y socialdatax-skills@latest auth login`,按隐藏输入提示填写,不主动要求其发到对话。Key 保存到当前执行用户主目录的 `.config/socialdatax/credentials.json`;Unix 权限 600,Windows 当前用户私有 ACL。不要在回复、输出日志、截图、命令示例、Skill 或版本库再次展示 Key;Agent 不读取或展示凭证文件内容。
- CLI 每次调用都会读取该私有用户文件,不依赖客户端加载 Shell 配置,不需要改客户端或系统设置。非空 `SOCIALDATAX_API_KEY` 优先,既有旧兼容变量其次,最后读取文件;`source=environment` 时保存文件不会覆盖生效的环境凭证;保存成功不等于本次所选 Key 已生效。CLI 提示不同 Key 冲突时,先说明实际来源,不擅自删除或改写既有环境配置。鉴权失败先核对实际来源和账号,再由用户本地修正,保留其它配置。
- 完成保存后,在实际业务命令的同一执行环境重新运行 `auth status`。false 时停止业务调用,核对操作系统用户、主目录与文件访问权限;true 只说明凭证可读,鉴权与结果仍以原业务请求返回为准,不为配置检查额外调用收费接口。
- 远程或容器中的 CLI 读取远程或容器用户的配置,本机保存不会自动传入。具备持久主目

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The authentication check executes socialdatax-skills@latest through npx -y, allowing a mutable registry release to run without a version-specific review or installation confirmation. A compromised release could access credentials and other files with the agent's permissions, even though the intended command only reports authentication status.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The interactive login command also resolves the mutable @latest release and executes it automatically. Hidden input and private credential-file permissions do not prevent a compromised CLI from collecting the key during login.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The stdin login command executes an unpinned @latest package while directly supplying an API key. If a future release or publisher account is compromised, that code could steal the key; passing it through stdin protects against argument exposure but not against the receiving program.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The XHS search example runs a mutable npm release through npx -y rather than a pinned dependency. Although this is a direct CLI rather than an MCP server, the supply-chain concern is valid: compromised package code would execute with access to the agent's environment and potentially its API key.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The XHS detail command automatically executes socialdatax-skills@latest, whose resolved code can change independently of this skill. A malicious replacement could read secrets or alter local files despite the documented read-only business operation.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The XHS comments command uses an unpinned @latest package with automatic execution. A compromised release could access the configured API key and local resources; the limits on comment retrieval do not mitigate that execution risk.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Douyin search command executes a mutable npm release without a fixed reviewed version. If that release is compromised, its code runs with the agent's privileges and could expose credentials or local data, regardless of the intended read-only search.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Douyin detail example fetches and executes socialdatax-skills@latest through npx -y. This introduces the same mutable-dependency execution risk as the other examples, with potentially sensitive credentials available to a compromised release.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Douyin comments command automatically runs an unpinned npm release. Compromise of the package's release channel could turn this retrieval step into arbitrary code execution under the current user, including access to the service credential.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Tieba search command executes @latest rather than a fixed audited release. The skill's read-only data-access intent does not prevent compromised package code from accessing credentials or modifying the execution environment.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Tieba detail example resolves a mutable npm version and executes it automatically. A malicious future release could run with the agent's permissions and access the API key, even though the requested operation only retrieves a post.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx -y socialdatax-skills'.

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

The Tieba comments command repeats the unpinned npx execution pattern. A compromised release could execute arbitrary code with the current user's privileges; pagination and sample-size limits only constrain legitimate application behavior.

Content

No source excerpt is available for this finding.