iblai-api-agent-chat
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to a vendor-hosted MCP server at https://asgi.data.iblai.app/mcp/agent-chat/. This is the intended endpoint for facilitating live chat functionality with deployed agents.
- [COMMAND_EXECUTION]: Provides setup instructions using the claude mcp add command. This allows the user to register the vendor's remote service as an MCP tool provider within their local environment.
- [INDIRECT_PROMPT_INJECTION]: The skill creates a vector for ingesting data from a remote agent.
- Ingestion points: Data entering the environment via the MCP server's chat tools.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the remote agent's responses.
- Capability inventory: Registering the MCP server grants the assistant new tools to send and receive messages.
- Sanitization: No sanitization of the remote agent's output is described in the setup process.
Audit Metadata