Security audit

商品评价分析

Security checks for vulnerabilities and agentic risk

Overview

The skill provides clearly scoped product-review analysis, with ordinary cautions about downloaded software and sending product information to SocialDataX.

Use existing review text for offline analysis. Online mode sends product URLs or share text to SocialDataX and may consume paid credits; remove unrelated personal details. Consider a reviewed, fixed CLI version and run it with only the access it needs.

A.I.G

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

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

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

Medium

Category: MCP Rug Pull
Confidence: 98% confidence
Finding: The product-detail command uses npx -y socialdatax-skills@latest, which can download and execute a changing package version without an installation confirmation. A compromised publisher account or malicious release could execute code with the agent's permissions and access SOCIALDATAX_API_KEY or other accessible data; the skill's read-only business functionality does not constrain package execution.
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 product-reviews command executes the mutable socialdatax-skills@latest package through npx -y. This is a CLI execution risk rather than specifically an MCP server configuration issue: a compromised future release could steal the API key or modify accessible local files despite the requested operation being read-only.
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 product-review-replies command repeats the automatic execution of an unpinned package using @latest. Although reply retrieval is optional, invoking it exposes the agent environment to whichever release the mutable tag resolves to, including a potentially 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: 98% confidence
Finding: The nested-reply command also uses npx -y socialdatax-skills@latest, allowing newly published package code to execute without a stable reviewed version. Optional use reduces exposure frequency but does not reduce the privileges available to malicious package code if the command runs.
Content: No source excerpt is available for this finding.

Online mode sends user-provided product URLs or complete share text to SocialDataX without a data-sharing warning.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: The workflow instructs the agent to submit the supplied URL or full share text through the SocialDataX product-detail API. Although the skill identifies the provider and describes online access, it does not warn that this user-provided content leaves the local environment; complete share text may include information beyond the product URL.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.