marketing-pipeline-share-automation
Warn
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to clone source code from a personal GitHub repository (
github.com/pennydinh/marketing-pineline-share.git) that is not associated with the skill's author or a trusted organization. - [COMMAND_EXECUTION]: The installation and usage instructions involve running package managers (
npm install) and execution scripts (npm run render,npm run dev) on the code downloaded from the external repository. - [INDIRECT_PROMPT_INJECTION]: The skill automatically crawls content from external websites (TechCrunch, X/Twitter, LinkedIn) and interpolates that data into prompts for LLMs (Claude and GPT-4), creating a surface for malicious data to influence agent behavior.
- Ingestion points: Research crawler module in
src/lib/crawlerthat fetches data from various social and news platforms. - Boundary markers: The prompt templates provided in the documentation do not implement delimiters or safety instructions to distinguish between system instructions and crawled research data.
- Capability inventory: The system has the ability to write files (video rendering via Remotion), perform network requests (crawling and API calls), and publish content to social media platforms.
- Sanitization: There is no evidence of sanitization, filtering, or validation performed on the crawled data before it is processed by the AI models.
Audit Metadata