marketingdashboard-financial-cockpit
Fail
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to clone a repository from an untrusted GitHub account (
theBigGavin/marketingdashboard). As this account is not recognized as a trusted organization or well-known service, it introduces a supply chain risk. - [REMOTE_CODE_EXECUTION]: The skill requires the user to execute code from the downloaded untrusted repository using commands like
npm install,npm run dev, andnpm start. This pattern of downloading and then executing arbitrary code from an unknown source is a high-severity security concern. - [INDIRECT_PROMPT_INJECTION]: The skill features a financial news aggregator (
/api/news) and industry chain parsing (/api/chain-parse) that ingest data from external, third-party sources. - Ingestion points: External data enters the dashboard via
/api/news,/api/quotes, and user-supplied text to/api/chain-parse. - Boundary markers: There are no mentioned boundary markers or instructions for the agent to ignore embedded commands within the processed news or data fields.
- Capability inventory: While primarily a dashboard, the underlying server implementation (
server/index.cjs) performs network operations and environment variable reads. - Sanitization: The documentation does not specify any sanitization, filtering, or escaping of the external content before it is rendered in the UI or processed by the AI features.
Recommendations
- AI detected serious security threats
Audit Metadata