Founder Article

• Outcome: Review
• Latest audit: May 29, 2026, 9:29 PM
• Version:
  1.0.0

Overview

This skill is a coherent article generator, but its optional publishing flow can create GitHub repositories and GitHub Pages sites without clear consent, visibility, or cleanup safeguards.

Use this for local drafting and PDF generation with generated content you are comfortable processing locally. Before allowing any GitHub publishing, confirm the exact files, repository name, repository visibility, Pages URL, and whether the content will be public; avoid publishing confidential business strategy or investor material without a manual review.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 78% confidence
• Finding: The skill advertises content generation but static analysis indicates file_read capability without any declared permission boundary. Undeclared file access is risky because it can pull in local content unrelated to the user's request, including sensitive workspace files, and the absence of an explicit permission model reduces user awareness and reviewability.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 88% confidence
• Finding: There is a significant description-behavior mismatch: the skill claims to generate polished business articles, but analysis suggests additional packaging behavior and reliance on local browser binaries for PDF rendering. This is dangerous because users may consent to a benign-seeming content task while the skill performs unexpected local operations, increasing the chance of unauthorized packaging, execution path abuse, or data exposure through auxiliary tooling.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The MRD explicitly describes a flow that creates or updates a GitHub repository and automatically deploys to GitHub Pages, but it does not mention any requirement for explicit user confirmation, scoping, or safeguards before publishing externally. In an agent skill, automatic external publication can expose sensitive drafts, overwrite repository contents, or cause unintended public release, making this a real security-relevant design issue even though it appears product-driven rather than malicious.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The README explicitly states that the skill will create a GitHub repository, enable GitHub Pages, and return a shareable URL, but it does not warn users that their generated content may be published externally and become publicly accessible. In a content-generation skill aimed at founders and investors—who may include confidential strategy, market analysis, or proprietary business information—this omission creates a real risk of unintended data disclosure.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill describes optional GitHub repository creation and GitHub Pages publishing without a prominent warning that content may be made public externally. This is dangerous because users may unintentionally publish proprietary business analysis, internal strategy, or personal data to a public repository or website, creating immediate confidentiality and reputational risk.

VirusTotal

64/64 vendors flagged this skill as clean.