Security audit

国际平台组测试合集

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent JD logistics lookup skill, but it uses token-authenticated HTTPS requests while disabling certificate checks, so it should be reviewed before use.

Install only if you trust the publisher and are authorized to query and disclose the referenced JD logistics data. Use a narrowly scoped token, avoid sharing or committing it, and patch or require an updated version that removes rejectUnauthorized: false before sending real credentials or business data.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Behavioral ASTexec() Call, eval() Call, Dynamic Import
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: The request explicitly sets rejectUnauthorized: false, which disables TLS certificate validation and allows man-in-the-middle interception or spoofing of the outbound API connection. In a logistics data-query skill that sends an authentication token in headers, this can expose credentials and permit tampering with returned operational data.

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: The script sets rejectUnauthorized: false, which disables TLS certificate validation for the HTTPS request. That makes the connection vulnerable to man-in-the-middle interception or spoofing, allowing an attacker on the network path to read or alter tracking queries and responses, including any token-based authentication sent in headers.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The README instructs users to place an API credential into persistent user environment settings and even shows a credential-like value, but provides no warning about secure secret handling, rotation, or exposure through shell history, screenshots, shared profiles, or local process inspection. In a logistics skill that accesses operational data, mishandled credentials could allow unauthorized access to shipment tracking or supply-chain metrics.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The skill explicitly instructs the agent to check environment variables before querying, which implies access to local sensitive configuration such as API keys, tokens, or internal endpoints. Without guardrails stating that secrets must never be surfaced, transformed into outputs, or exposed through errors, an agent could inadvertently read or disclose sensitive runtime configuration while fulfilling user requests.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The markdown directs the agent to execute local Node.js commands using user-derived parameters, but does not disclose this execution behavior or define safety controls around command construction. In agent environments, hidden command execution increases the risk of unsafe local actions, accidental shell injection in downstream implementations, and user-unexpected access to local resources.

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The skill is designed to retrieve and display customer, warehouse, owner, and product-related operational data, yet provides no privacy, authorization, or data-minimization guidance. In a logistics context, these identifiers and performance records may be commercially sensitive, and exposing them without explicit access control expectations can enable unintended disclosure to unauthorized users.

Missing User Warnings

High

Confidence: 99% confidence
Finding: Disabling TLS validation without warning or justification makes the HTTPS protection ineffective against forged certificates. Because this skill accesses an external logistics API and includes a token header, the context increases risk: an attacker on the network path could steal the token, alter responses, or mislead downstream users with falsified logistics metrics.

Missing User Warnings

High

Confidence: 99% confidence
Finding: Disabling TLS certificate validation with rejectUnauthorized: false allows man-in-the-middle interception or tampering of HTTPS traffic. Because this script sends an authentication token and likely retrieves logistics or operational data, an attacker on the network path could steal credentials or alter returned business data.

Static analysis

No suspicious patterns detected.