brand-mention-monitor
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches visualization libraries (D3.js and TopoJSON) and geographic data (world-atlas) from well-known content delivery networks including cdnjs, jsDelivr, and unpkg to render the triage console.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the web and social media to calculate sentiment and risk scores.
- Ingestion points: External posts, articles, and excerpts are ingested via
nimble searchresults. - Boundary markers: The skill relies on structured data processing rather than explicit prompt delimiters for the ingested content.
- Capability inventory: Employs file system operations (
Writeto~/.nimble/), agent orchestration (Agent), and user input requests (AskUserQuestion). - Sanitization: The implementation includes dedicated sanitization logic (HTML escaping and URL validation) within its output templates to mitigate risks associated with displaying untrusted external data.
Audit Metadata