ee-ai-toolkit

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 6:47 AM UTC
• Version: 1.0.0

Overview

This skill is a disclosed electrical-engineering learning toolkit with local example scripts and no evidence of hidden network access, exfiltration, persistence, or unsafe automatic execution.

Install only if you want a local electrical-engineering AI course/reference toolkit. Treat all calculation, protection, sizing, and optimization outputs as educational drafts that require independent engineering verification. Do not enter confidential prompts, credentials, resumes, job-search details, or sensitive meter data into the example scripts unless you are comfortable with local processing or local file output.

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

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• System Prompt Leakage: Direct Leakage, Indirect Extraction, Tool-Based Exfiltration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection

Findings (17)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 94% confidence

Finding

The declared description presents the skill as an electrical engineering AI toolkit focused on domain-specific engineering tasks and related example scripts. The supplied code does not perform electrical engineering calculations, analysis, visualization, optimization, or any domain-specific function. Instead, it collects arbitrary user input and saves it to a local text file. That is a materially different primary purpose and introduces an undeclared data logging/storage capability, so this is a clear mismatch.

Content

No source excerpt is available for this finding.

Direct Prompt Extraction

High

• Category: System Prompt Leakage
• Confidence: 85% confidence

Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · assets/python-scripts/script_061_power_system_prompt_pack_generator.py (reported line 9)May include surrounding context.

python

f"Calculate fault current in a {system_type} network.",
    f"Analyze voltage stability of a {system_type} system."
    ]
    return prompts

system = input("Enter system type: ")

Direct Prompt Extraction

High

• Category: System Prompt Leakage
• Confidence: 85% confidence

Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · references/source-digest.md (reported line 5365)May include surrounding context.

md

f"Calculate fault current in a {system_type} network.",
    f"Analyze voltage stability of a {system_type} system."
    ]
    return prompts

system = input("Enter system type: ")

Undeclared Tool Scope

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.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 89% confidence

Finding

This markdown file states that the 'Prompt Logger' stores prompts for reuse, which implies retention of user-supplied content. The description does not warn users that prompts may contain sensitive or proprietary information and will be persisted.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 84% confidence

Finding

The 'Prompt Usage Analyzer' is described as tracking frequency of prompt usage, which involves monitoring user activity. The file provides no disclosure that usage data may be collected, stored, or analyzed.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 81% confidence

Finding

Smart meter aggregation can involve household or facility consumption data that is operationally sensitive. The description presents this processing with no warning about handling potentially private energy-usage data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 87% confidence

Finding

Resume analysis necessarily processes personal career information and potentially contact or employment history data. The markdown description does not disclose that users may be providing sensitive personal information to the skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 85% confidence

Finding

Tracking job applications involves storing personal employment-seeking activity, which can be sensitive. The catalog entry does not warn users about persistence or privacy implications of this data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

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

Finding

The extracted script listing explicitly identifies a 'Prompt Logger' script, indicating prompts may be written to storage. The markdown file does not warn that logged prompts could include confidential or sensitive information.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 86% confidence

Finding

These entries describe analyzers and trackers for prompt usage and performance, which implies collection of behavioral metadata. No user-facing warning is provided about privacy, retention, or sensitivity of activity data.

Content

No source excerpt is available for this finding.

Ae4

Medium

• Category: analysis-evasion
• Confidence: 80% confidence

Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 84% confidence

Finding

The digest contains extensive AI-generated calculation, design, protection, and optimization guidance, plus executable example scripts, but the later modules and appendix do not consistently repeat strong safety/validation warnings near operational content. In an electrical-engineering context, users may overtrust generated outputs for design, protection settings, load flow, or equipment sizing, which can propagate unsafe real-world decisions.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 80% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/source-digest.md (reported line 1329)May include surrounding context.

md

Apply AI to assist in electrical design tasks

Generate calculation scripts automatically

Perform basic power system analysis

Missing User Warnings

Low

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

Finding

The script silently appends user-entered prompts to a local file without informing the user about retention or append behavior. This can unintentionally store sensitive data such as API keys, credentials, proprietary prompts, or personal information, creating a privacy and data-handling risk even though the behavior is simple and likely intended for convenience.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

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

Finding

The rewritten prompt always prepends 'Act as an electrical engineer,' which imposes a fixed domain/persona regardless of the user's intent. This is a natural-language policy concern because it hard-codes a specific mode of interaction without offering user opt-in or alternatives.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 78% confidence

Finding

A LinkedIn optimization tool would typically process profile text and career information. The markdown entry lacks any notice that personal profile data may be analyzed or retained.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.