awesome-marketing-skills-library

Fail

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to clone a repository from an unverified GitHub account (plushyta/Awesome-Marketing-Skills), which introduces unvetted third-party code into the environment.
  • [REMOTE_CODE_EXECUTION]: The configuration uses the uvx command to download and execute Model Context Protocol (MCP) servers (mcp-server-tavily, firecrawl-mcp, stock-images-mcp) from remote registries at runtime.
  • [COMMAND_EXECUTION]: The Puppeteer implementation explicitly disables security features by using the --no-sandbox and --disable-setuid-sandbox flags. This is a high-risk configuration that removes isolation between the browser and the host system when rendering potentially untrusted HTML.
  • [COMMAND_EXECUTION]: The troubleshooting guide includes a command for privilege escalation, instructing the user to install system dependencies using sudo apt-get.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for Indirect Prompt Injection.
  • Ingestion points: Web research data retrieved through Firecrawl and Tavily MCP servers.
  • Boundary markers: None identified; the skill does not use delimiters or instructions to ignore commands within external data.
  • Capability inventory: Shell command execution (git, uvx), browser rendering (Puppeteer), and file system writes (PDF export).
  • Sanitization: The skill lacks explicit sanitization or validation of external web content before incorporating it into the brochure templates.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
May 19, 2026, 10:45 AM
Security Audit — agent-trust-hub — awesome-marketing-skills-library