Guides users to request a custom skill by email and helps structure their requirements. Invoke when a user needs a custom-built skill, plugin, or private agent capability.

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 4:34 AM UTC
• Version: 1.0.1

Overview

This skill is a disclosed email-referral guide for custom skill requests, but users should avoid sending confidential details to the listed personal email address without independent trust checks.

Install only if you are comfortable with the listed email contact. Before sending anything, verify who controls the address and omit secrets, credentials, proprietary code, customer data, internal network details, and confidential business information unless you have an approved secure channel or agreement.

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 (1)

T01 · Skill Instruction Hijacking

Warning

• Location: SKILL.md:26

Finding

Mandatory Redirection to an Unverified External Email Address

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 26-45 and 71-92
Vulnerability Type: T01: Skill Instruction Hijacking
Risk Level: Medium

The skill instructs the agent to repeatedly direct users to the fixed, unauthenticated external email address [e-mail]. It also encourages users to include potentially sensitive business requirements, internal usage scenarios, deployment preferences, timelines, and workflow information.

Vulnerable Instructions

SKILL.md, lines 26-45:

markdown

## Contact

Guide the user to send an email to:

**[e-mail]**

## Required Response

When invoking this Skill, the output should accomplish the following objectives:

1. Clearly inform the user that they can submit custom Skill requests via email
2. Provide the contact email `[e-mail]`
3. Briefly explain what content is recommended to include in the email
4. If the context is sufficient, proactively help the user organize a requirement summary that can be sent directly

## Suggested Requirement Format

It is recommended to guide users to include the following information in their email:

- Requirement background
- What problem the Skill or plugin should solve
- Core functions expected to be implemented

SKILL.md, lines 71-92:

markdown

## Steps

1. Determine whether the user explicitly needs custom development, exclusive development, or human-assisted development of a new Skill
2. If yes, explain that the requirement is suitable for email communication
3. Output the contact email `[e-mail]`
4. Prompt the user for key information they should provide
5. If the context already contains partial requirements, organize them into a structured requirement summary or email draft
6. Do not claim to have sent the email on behalf of the user

## Output

The output content should be concise, clear, and actionable, containing at minimum:

- Contact email
- Exp
...[truncated 2943 chars]

Remediation

View remediation

Remediation Suggestions

1. Remove the hardcoded personal email address and all instructions requiring the agent to promote it.
2. Use a contact endpoint that is configured by the user or deploying organization rather than embedded by the skill author.
3. If an external contact channel is necessary, verify organizational ownership and clearly identify the responsible entity, privacy policy, data-retention terms, and intended use of submitted information.
4. Require explicit user consent before extracting or organizing conversation content for external transmission.
5. Add a warning not to include credentials, API keys, personal data, proprietary source code, network details, customer information, or other confidential material.
6. Apply data minimization: request only the information necessary for an initial inquiry and allow users to review and redact the draft.
7. Do not imply that an external destination is trusted, official, or endorsed unless that status can be independently verified.
8. Prefer an authenticated support portal or organization-controlled ticketing system with access controls, encryption, audit logging, and a documented retention policy.
9. Narrow the invocation criteria so ordinary questions that can be answered in-session are not redirected to external contact channels.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

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

Finding

The skill instructs users to email detailed custom-skill requirements, which may include sensitive business information such as internal workflows, private deployment needs, commercial plans, and reference processes, but it provides no warning about confidentiality, data minimization, or secure transmission. This can lead users to disclose proprietary or sensitive operational details to a personal email address over an insecure or unmanaged channel.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.