The Agent Skills Directory

[SAFE]: The skill acts as a standard integration library for the iGrant.io platform. No malicious command execution, data exfiltration, or persistence mechanisms were detected.\n- [SAFE]: The documentation and code samples correctly guide the agent to handle credentials securely, specifically emphasizing that API keys should be stored server-side and not exposed in the browser environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill handles data from external API endpoints and Server-Sent Events (SSE), which is a necessary component of its functionality.\n
Ingestion points: Data is received via network responses in owsClient.ts and SSE events in sseClient.ts.\n
Boundary markers: None are present; the data is handled as structured JSON by the client logic.\n
Capability inventory: The skill is limited to browser-based network requests (fetch) and UI rendering; it has no access to sensitive file systems or system-level execution.\n
Sanitization: Incoming data is parsed using JSON.parse and mapped to strictly defined TypeScript interfaces in types.ts.

igrantio-frontend-client