marketing-pipeline-share-ai-content

Warn

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to clone and execute code from an unverified external repository at https://github.com/pennydinh/marketing-pineline-share.git. This source does not match the skill author's verified infrastructure or any recognized trusted organization, posing a supply chain risk.
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by processing data from external, untrusted web sources and inserting it directly into LLM prompts.
  • Ingestion points: The conductResearch module in lib/research/scraper.ts fetches data from TechCrunch, a16z, Twitter, and LinkedIn.
  • Boundary markers: The buildPrompt function in lib/ai/generator.ts embeds this raw research data directly into the prompt using simple string concatenation without boundary markers or instructions to ignore embedded commands.
  • Capability inventory: The processed data flows into powerful LLM models (Claude 3.5 Sonnet and GPT-4) used for content generation.
  • Sanitization: The implementation lacks sanitization or validation of the external content before it is processed by the AI models.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 26, 2026, 12:58 AM
Security Audit — agent-trust-hub — marketing-pipeline-share-ai-content