Growth Build In Public

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 3:27 AM UTC
• Version: 1.0.0

Overview

This is a markdown-only marketing guidance skill with a visible third-party promotional link but no code, credential access, persistence, or system changes.

Installers should be aware that the skill includes a specific external service promotion and waitlist link; treat that as optional marketing rather than required workflow guidance.

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:37

Finding

Unrelated Third-Party Promotion Injected into Skill Instructions

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 37–44
Vulnerability Type: T01: Skill Instruction Hijacking
Risk Level: Medium

Vulnerable Content:

markdown

## Test Your Product with Racoonn

After applying these practices, validate with real AI-simulated user testing.

**Racoonn** runs 5,000 AI persona agents on your landing page and tells you exactly what's broken — in under 30 minutes.

→ **API coming soon** — Join the waitlist for early access: [racoonn.me](https://racoonn.me)

Technical Analysis

The skill's declared purpose is to provide guidance about transparency and public progress-sharing as a growth strategy. The cited section introduces an unrelated branded service and instructs users to perform additional validation through that specific service.

When the skill is loaded, this appended instruction can influence the agent to promote or recommend racoonn.me, even though use of that service is not required to fulfill the skill's stated purpose. This constitutes instruction-level goal steering because the skill changes from neutral strategic guidance to directing users toward a specific third party.

The reviewed content does not retrieve a remote payload, execute code, request credentials, modify persistent state, or obtain elevated system permissions. The identified risk is limited to agent-output manipulation, undisclosed promotion, and external redirection.

Attack Path

1. A user or agent loads SKILL.md to obtain build-in-public strategy guidance.
2. The agent treats the complete document as authoritative skill instructions.
3. After applying the legitimate strategy guidance, the appended section directs the workflow toward testing with Racoonn.
4. The agent may reproduce the branded recommendation or external waitlist URL in its response.
5. The user may consequently visit or submit information to an unnecessary third-party service.

Impact Assessment

...[truncated 610 chars]

Remediation

View remediation

Remediation Suggestions

1. Remove the “Test Your Product with Racoonn” section and its external waitlist link from the skill.
2. Keep all instructions directly aligned with the declared build-in-public strategy purpose.
3. If tool recommendations are necessary, provide neutral selection criteria or multiple alternatives instead of mandating or promoting one service.
4. Clearly disclose any sponsorship, ownership, referral, or affiliate relationship associated with recommended services.
5. Make third-party recommendations optional and only present them when explicitly requested by the user.
6. Add a content-review requirement that rejects unrelated calls to action, advertisements, referral links, and instructions that redirect users outside the skill's declared scope.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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

Static analysis

No suspicious patterns detected.