marketingskills-ai-agent-collection

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted user data—such as product names, audience personas, and value propositions—to generate marketing copy, email sequences, and SEO audits.
  • Ingestion points: Prompts requesting product marketing context in SKILL.md.
  • Capability inventory: The agent is instructed to generate files and content (copywriting, emails, SEO reports) based on these inputs.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the input data are present in the provided snippets.
  • Sanitization: No explicit sanitization or filtering of the processed marketing data is mentioned.
  • [EXTERNAL_DOWNLOADS]: The installation section provides instructions for fetching content from an external source.
  • Evidence: The skill recommends using npx skills coreyhaines31/marketingskills and git clone https://github.com/coreyhaines31/marketingskills.git to install the marketing skill collection.
  • [COMMAND_EXECUTION]: The documentation includes several shell commands for installation and maintenance.
  • Evidence: Commands such as npx skills, git clone, cp -r, and git submodule are provided in the documentation as setup steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:35 PM