tanstack-ai-mcp
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the use of
stdioTransportinSKILL.md, which allows the agent to interface with local MCP servers by spawning child processes and executing shell commands such asnpx. - [EXTERNAL_DOWNLOADS]: The skill facilitates network communication with external MCP servers through HTTP and SSE transports, as described in the client configuration examples.
- [PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection by processing data (tools, resources, and prompts) from external MCP servers.
- Ingestion points: Data enters the agent's context through
client.tools(),client.readResource(), andclient.prompts()methods defined inSKILL.md. - Boundary markers: The skill does not define specific delimiters; it relies on the underlying
chat()loop implementation for message framing. - Capability inventory: The skill has the capability to execute local commands (via
stdioTransport) and perform network requests (via HTTP/SSE). - Sanitization: There is no explicit documentation of sanitization or filtering of the content retrieved from external MCP servers before it is processed by the agent.
Audit Metadata