iblai-agent-chat
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configures a connection to a hosted MCP server at
https://asgi.data.iblai.app/mcp/agent-chat/. This endpoint is used for streaming agent responses and tool interactions with the ibl.ai platform. - [COMMAND_EXECUTION]: Provides instructions for the user to execute the
claude mcp addcommand. This is used to register the vendor's MCP server and configure headers for authentication and agent identification. - [PROMPT_INJECTION]: The skill establishes a channel for indirect prompt injection by processing responses from an external agent.
- Ingestion points: External agent responses received via the
iblai-agent-chatMCP server. - Boundary markers: None explicitly defined in the skill instructions to separate external agent output from system instructions.
- Capability inventory: The agent utilizes tools provided by the
iblai-agent-chatMCP server to facilitate the conversation. - Sanitization: No specific sanitization or validation logic for the ingested agent responses is described.
Audit Metadata