NYC

reactive-programming

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill demonstrates patterns for ingesting external data through WebSocket messages and user search inputs. While the examples only involve low-risk capabilities like UI updates and local state management, these entry points represent a potential surface for indirect injection if the agent uses this logic to influence higher-privilege actions.
  • Ingestion points: searchInput (DOM event), event.data (WebSocket).
  • Boundary markers: None present in code examples.
  • Capability inventory: console.log, displayResults (UI), StateManager (local state).
  • Sanitization: No explicit sanitization or validation of the ingested strings is shown.
  • [Data Exposure & Exfiltration] (SAFE): No access to sensitive file paths (~/.ssh, etc.) or hardcoded credentials. All network operations use relative API paths or generic placeholders.
  • [External Downloads] (SAFE): The skill references the standard and highly trusted 'rxjs' library. No untrusted third-party dependencies or remote script execution (curl|bash) were detected.
Recommendations
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 10:23 PM