Security audit

China Electronic Components Factory

Security checks for vulnerabilities and agentic risk

Overview

This is a mostly static China electronics sourcing guide, but it needs Review because it says it has no individual factory contacts while bundling a direct supplier contact directory.

Install only if you are comfortable with a sourcing skill that may surface specific supplier names and email contacts. Treat the bundled supplier directory as unverified lead-generation data, confirm vendors independently, and avoid using it for automated outreach without separate user approval and compliance review.

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

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 88% confidence
Finding: The skill is invocable and its implementation reportedly includes file_read capability, but the manifest does not declare any tool scope such as permissions or allowed-tools. This creates a policy and transparency gap: the runtime may permit filesystem access that users or reviewers cannot see from the manifest, increasing the risk of unintended local file access or data exposure if the code reads beyond the intended data files.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The file includes a sourcing-ready factory directory with direct supplier email contacts, capacities, and certifications, which goes beyond neutral market research and materially enables procurement targeting. In an agent skill, this can be used to steer users toward specific vendors, facilitate unsolicited outreach, or operationalize supply-chain targeting without additional vetting or user safeguards.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The manifest explicitly scopes the guide to semiconductor, passive component, PCB, connector, and sensor manufacturing clusters. Entries for BOE, Tianma, Innolux, and CSOT are display/panel manufacturers, which are a different product category than the declared component domains.

Vague Triggers

Low

Confidence: 86% confidence
Finding: This JSON manifest-like file contains extensive content data but no explicit activation conditions, trigger phrases, or exclusion criteria for when the skill should run. For manifest-scope files, missing trigger specificity can make invocation behavior ambiguous because there is no documented boundary between intended and unintended use.

Natural-Language Policy Violations

Low

Confidence: 78% confidence
Finding: The top-level docstring frames the skill as specifically for 'China electronic components industry intelligence' and sourcing guidance, which imposes a fixed geographic/locale scope in natural language. While this may be intentional, the file does not explicitly present this as a user-selected regional specialization or document why the locale restriction is required.

Static analysis

No suspicious patterns detected.