Back to skill

Security audit

xhs-dpt

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because scanners report insecure network settings and collection capabilities that extend beyond its stated engagement-metrics purpose.

Consider the reported credential exposure, optional TLS verification bypass, and broader collection capabilities before installing. Skillspector coverage was partial, with 2 of 7 components fully scanned and 5 partially inspected, so some behavior remains unverified.

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 (1)

T09 · Insecure Skill Coding Practices

Warning
Location
src/config.js
Finding

Default HTTP transport exposes authentication credentials

Content
View full analysis

Impact

An intermediary can capture and replay bearer credentials to access account data and perform authenticated operations, including consuming collection credits. An active intermediary can also tamper with returned results.

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (7)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied command explicitly requests logout. It does not collect note metrics, process note links, or export CSV results. Although authentication management could support the declared service, the supplied chunk performs only logout, an undeclared operation materially different from the described collection workflow. The CLI implementation is not provided, so its underlying effects cannot be verified.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The generic task and export methods are consistent with the declared service, and authentication and networking are supporting implementation details. However, the client also exposes concrete create, step, finish, and retrieval methods for multiple collection workflows outside the declared scope. These include different data types and other platforms, making the supplied code materially broader than the description. These are callable API wrappers; the chunk does not demonstrate that the extra workflows execute automatically. CSV formatting, xsec_token enforcement, and pricing are not established by this client alone and may be implemented elsewhere.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The activation description includes the generic trigger “数据采集” (“data collection”) without requiring Xiaohongshu context.

Medium
Category
Trigger Abuse
Confidence
96% confidence
Finding

The description explicitly permits activation on the generic phrase “data collection,” allowing unrelated requests to be routed into this specialized third-party paid-service workflow. This creates a routing and consent risk, although activation alone does not demonstrate data disclosure or a completed charge.

Content

No source excerpt is available for this finding.

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
99% confidence
Finding

When called, allowInsecureTls disables TLS certificate verification process-wide, allowing a network attacker to impersonate HTTPS services and intercept bearer tokens or alter responses. This is an optional insecure mode rather than a demonstrated default, but its global scope affects subsequent TLS connections beyond the intended endpoint, and warnings do not restore authentication.

Content

Scanner excerpt · src/client.js (reported line 75)May include surrounding context.

js
/** 忽略 HTTPS 证书校验 (服务端自签/域名不匹配证书时由 insecure 配置开启)。进程级。 */
function allowInsecureTls() {
  process.env.NODE_TLS_REJECT_UNAUTHORIZED = '0';
}

/** 业务错误: code 为后端错误码 (1001/1002/3001/...) */

The client exposes profile, comment-content, and cross-platform collection capabilities beyond Xiaohongshu note interaction counts.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The manifest limits this skill to collecting aggregate interaction metrics for supplied Xiaohongshu note links and exporting those results. The exported ApiClient additionally implements profile-note discovery (L178–L205), Xiaohongshu comment-content collection (L257–L282), Douyin danmaku collection (L207–L233), and comment collection across numerous other platforms (L284–L519). These callable session-creation, stepping, and retrieval methods enable collection workflows beyond the declared purpose, although this file alone does not establish that they are invoked automatically.

Content

No source excerpt is available for this finding.

Mandatory verbatim registration guidance forces Chinese output without a user language choice.

Low
Category
Output Handling
Confidence
93% confidence
Finding

The skill requires the agent to send the registration guidance unchanged, and that guidance is written in Chinese. This mandates Chinese user-facing output regardless of the user's language, without an opt-in or a documented reason the wording must remain untranslated.

Content

No source excerpt is available for this finding.