T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:36- Finding
Customer PII and Sensitive Feedback May Be Exposed in Shared Roadmap Documents
- Content
View full analysis
- Remediation
View remediation
Security audit
Feedback Roadmap
Security checks for vulnerabilities and agentic risk
Overview
The skill is a coherent feedback-roadmap guide, but it pushes users to collect and share raw customer communications without adequate privacy safeguards.
Review before installing if you will use real customer data. Only use sanitized, authorized feedback; remove or hash identifiers, redact quotes, avoid sensitive transcript content, restrict report access, and define retention/deletion rules before sharing any roadmap.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
Findings (1)
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The skill explicitly instructs analysis of customer feedback sources such as reviews, surveys, support transcripts, return reasons, and social DMs, which commonly contain personal or sensitive data. It provides operational guidance for collecting and normalizing that data but omits privacy, minimization, redaction, consent, retention, or access-control safeguards, creating a realistic risk of over-collection or improper handling of customer information.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The workflow directly tells users to pull raw feedback from support chat/email transcripts and social DMs into a normalized table with verbatim content and customer value fields. Because these sources frequently contain names, contact details, order data, and other sensitive context, the lack of explicit warnings or sanitization steps increases the chance that users will centralize sensitive data inappropriately and expose it to broader audiences or downstream systems.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The template explicitly instructs users to keep verbatim customer quotes and aggregates support tickets, social DMs, and survey free text, but provides no warning to remove or mask personally identifiable information or other sensitive data before sharing. In a workflow designed for broad circulation, this increases the likelihood of exposing names, contact details, order data, health/payment details, or other sensitive content copied from raw feedback into a report.
- Content
Static analysis
No suspicious patterns detected.