Back to skill

Security audit

快手评论分析 SocialDataX 评论洞察

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because its instructions repeatedly run a changeable, unpinned CLI package, creating a supply-chain risk; the scan did not find clear evidence of malicious intent.

Consider postponing installation until the CLI package is pinned to a reviewed version and the missing references are clarified. If you proceed, review the login and credential-handling instructions carefully; the partial scan leaves some evidence 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (10)

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 26)May include surrounding context.

md
ion and authorized configuration, send it as stdin to `npx -y socialdatax-skills@latest auth login --stdin` for persistent CLI storage; do not request duplicate manual entry, pass it as a CLI argument, or write it to a temporary script file. Otherwise direct the user to the official URL above and hidden input with `npx -y socialdatax-skills@latest auth login` in their local terminal, without asking them to send it in chat. The CLI saves it under the execution user's home at `.config/socialdatax/credentials.json`, Unix mode 600 or a private current-user ACL on Windows. Do not repeat the Key in replies, output logs, screenshots, command examples, Skill files or version control. The Agent must not read or display credential-file content.
- Each CLI invocation reads the private user file without depending on a client loading shell configuration or changing client/system settings. A nonempty `SOCIALDATAX_API_KEY` has priority, then the existing legacy variable, then the file. Saving a file

Activation criteria include generic analysis requests without requiring a Kuaishou context.

Medium
Category
Trigger Abuse
Confidence
92% confidence
Finding

The invocation instruction includes standalone requests for audience feedback, sentiment themes, pain points, FAQ extraction, and discussion summaries. Unlike the Kuaishou-specific triggers in the same sentence, these alternatives could activate the skill for unrelated platforms or user-provided text, and no explicit exclusion limits them to Kuaishou comments.

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
98% confidence
Finding

The authentication-status instruction executes socialdatax-skills@latest through npx with automatic confirmation, allowing the executed package version to change without review. A compromised future release could execute with the agent's permissions and access available credentials, even though the intended command only reports credential readability; this is a CLI supply-chain issue, not an MCP-server configuration issue.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 26)May include surrounding context.

md
an explicit Key change. When the user has already supplied a Key in the installation prompt or current conversation and authorized configuration, send it as stdin to `npx -y socialdatax-skills@latest auth login --stdin` for persistent CLI storage; do not request duplicate manual entry, pass it as a CLI argument, or write it to a temporary script file. Otherwise direct the user to the official URL above and hidden input with `npx -y socialdatax-skills@latest auth login` in their local terminal, without asking them to send it in chat. The CLI saves it under the execution user's home at `.config/socialdatax/credentials.json`, Unix mode 600 or a private current-user ACL on Windows. Do not repeat the Key in replies, output logs, screenshots, command examples, Skill files or version control. The Agent must not read or display credential-file content.
- Each CLI invocation reads the private user file without depending on a client loading shell configuration or changing client/system settings

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

Medium
Category
MCP Rug Pull
Confidence
98% confidence
Finding

The interactive login example also runs socialdatax-skills@latest without fixing the executable version. Hidden input protects against visible disclosure but cannot stop a malicious package release from capturing the entered key or accessing other resources available to the process.

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
98% confidence
Finding

The stdin-based login command executes the mutable socialdatax-skills@latest package immediately before supplying an API key to it. Stdin avoids command-line secret exposure but does not protect the key from a compromised package receiving it.

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
98% confidence
Finding

The photo-ID comments command invokes a mutable @latest npm release with automatic confirmation. Although the intended operation is read-only, a compromised CLI package would run with the execution user's permissions and could access the API key or other accessible data.

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
98% confidence
Finding

The URL-based comments example executes socialdatax-skills@latest rather than an immutable reviewed version. The legitimate purpose of fetching comments does not constrain what a compromised npm release could do with the process's credentials and permissions.

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
98% confidence
Finding

The replies command repeats the unpinned @latest execution pattern, so future package changes can silently alter the code run by the agent. This shares the same supply-chain root cause as the other CLI examples rather than demonstrating a separate malicious behavior.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx availability'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.