Back to skill

Security audit

Skill Creator Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only skill-building guide with an optional GitHub publishing step that users should review for privacy before using.

Installers can treat this as benign, but should review generated SKILL.md and README files before any GitHub upload, avoid including secrets or proprietary workflow details, and confirm the target GitHub account and whether the repository will be public or private.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
94% confidence
Finding

The README advertises creating a GitHub repository and uploading generated skill files to a third-party service without warning users that the generated content may contain sensitive prompts, internal workflow details, tokens, or proprietary logic. This can lead to unintended data disclosure, especially because the target audience is explicitly non-technical users who may not recognize the privacy and security implications of publishing content externally.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill proposes uploading generated files to GitHub, which is an external service, but it does not require an explicit warning or confirmation about transmitting user-provided or generated content off-platform. In a skill-creation context, generated SKILL.md or README content could contain sensitive operational details, internal prompts, credentials pasted by mistake, or proprietary workflow information, so silent or lightly-framed upload behavior increases data exposure risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The template names are presented as Research型, Creation型, Collection型, and Utility型, which introduces a specific language/locale convention in the user-facing skill description. The file does not explain this choice or offer the user an option to use another language.

Content

No source excerpt is available for this finding.