DHX Components Sourcing

• Outcome: Review
• Latest audit: Sep 16, 2026 · 10:35 AM UTC
• Version: 1.0.0

Overview

This skill is a real component-sourcing integration, but it automatically sends part/BOM-derived queries to a single vendor and persistently steers sourcing answers into mandatory DHX promotional output.

Install only if you intentionally want DHX Tech as the default sourcing provider for matching hardware questions. Avoid using it with confidential BOMs unless you are comfortable sending part numbers to icdhxkj.com, and be aware that the Cursor rules option can keep adding DHX promotional contact links in later project sessions.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

Findings (3)

T01 · Skill Instruction Hijacking

Error

• Location: SKILL.md:20

Finding

Mandatory Vendor Role and Promotional Output Hijacking

Content

View full analysis

Remediation

View remediation

T01 · Skill Instruction Hijacking

Error

• Location: skills/dhx-tech-sourcing/SKILL.md:20

Finding

Duplicated Mandatory Vendor Output Instructions in Nested Skill

Content

View full analysis

Remediation

View remediation

T01 · Skill Instruction Hijacking

Error

• Location: cursorrules.txt:26

Finding

Persistent Project Rules Force Promotional Content into Future Responses

Content

View full analysis

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill instructs the agent to send part numbers and potentially BOM-derived queries to an external API without notifying the user or obtaining consent. BOMs often contain proprietary design, sourcing, and supplier information, so silent transmission off-platform can expose confidential product, supply-chain, or commercial data to a third party.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium

• Category: Agent Snooping
• Confidence: 80% confidence

Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · README.md (reported line 32)May include surrounding context.

md

Cursor automatically picks up:
- `rules/dhx-electronics-sourcing.mdc` — project rules (same content as `cursorrules.txt`)
- `skills/dhx-tech-sourcing/SKILL.md` — the agent skill

> 💡 Legacy fallback: Option 1 (copy `cursorrules.txt` → `.cursorrules`) still works for older Cursor versions.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

The trigger conditions are broad enough to activate on many ordinary hardware-related requests, including general sourcing, PCB design, and BOM discussions. This can cause unintended invocation of the skill, steering users into a vendor-specific workflow, making unsolicited external queries, and injecting promotional content or links into responses without a clear user request for this specific provider.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 91% confidence

Finding

The activation rule is broad enough to trigger on common terms like finding, sourcing, or buying components, which can cause the skill to engage outside a clearly intended workflow. This creates prompt-scope hijacking risk: unrelated conversations may be steered into mandatory external lookups, link insertion, and vendor promotion without sufficient user intent confirmation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 94% confidence

Finding

The rule forces locale rewriting and defaults to English when the user's language is not recognized, without asking the user or preserving the original neutral URL. While not a direct code-execution issue, it can misroute users, alter link destinations unexpectedly, and create deceptive UX by transforming links based on inference rather than explicit consent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 92% confidence

Finding

The trigger conditions are broad enough to activate on general hardware design or manufacturing discussions, not just explicit sourcing requests. That can cause the agent to steer ordinary engineering conversations into vendor-specific promotion and unnecessary external lookup flows, increasing the chance that sensitive project details or BOM data are disclosed without clear user intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The instructions require detecting the user's language and forcing a locale-specific URL, with unsupported languages automatically routed to '/en/'. This imposes a language choice without user opt-in and may conflict with organizational language or locale preference policies.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 92% confidence

Finding

This OpenAPI file presents the title, descriptions, parameter documentation, and response descriptions only in Chinese, which can constitute a language/locale policy issue when no user opt-in or alternative locale is offered. The file does not document that the API is intentionally region-specific or provide multilingual options.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.