Cocreator Content Research

• Outcome: Pass
• Latest audit: May 28, 2026 at 11:17 AM UTC
• Version:
  1.0.2

Overview

This is a read-only social media research skill that transparently uses a ScrapeCreators API key, with a few privacy and installer cautions but no evidence of hidden posting, local data access, persistence, or destructive behavior.

Install only if you are comfortable giving the skill a ScrapeCreators API key and sending researched keywords, hashtags, and creator handles to ScrapeCreators. Use a limited API key or quota where possible, avoid passing the key on the command line, and review or install uv through a trusted package manager instead of blindly running the curl-to-shell installer.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• 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

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 81% confidence
• Finding: The skill uses environment secrets and makes external network calls, but it does not declare corresponding permissions or user-visible boundaries. That weakens platform-level governance and informed consent, making it easier for an agent to transmit user queries, handles, or API-backed results off-platform without clear review.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 88% confidence
• Finding: The skill is presented as 'pure intelligence gathering,' but the documented behavior includes broader collection features such as keyword/hashtag search, direct video playback URLs, and full raw profile responses. This mismatch can cause downstream agents or users to underestimate the sensitivity and breadth of data retrieval, increasing the chance of inappropriate data collection or disclosure.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The skill lacks a user-facing warning that queries, creator handles, and related research inputs are transmitted to an external API provider. In a research context, those inputs may reveal business strategy, client interests, or investigative targets, so silent transmission creates privacy and confidentiality risk.

External Script Fetching

Low

• Category: Supply Chain

Content

{
          "id": "uv-install",
          "kind": "bash",
          "script": "curl -LsSf https://astral.sh/uv/install.sh | sh",
          "bins": ["uv"],
          "label": "Install uv (cross-platform via bash)"
        }

• Confidence: 93% confidence
• Finding: curl -LsSf https://astral.sh/uv/install.sh | sh

VirusTotal

64/64 vendors flagged this skill as clean.