openapi-mcp

The materials explicitly state that no keys or environment variables are required, and there is no indication that the tool itself requests or stores sensitive credentials; if downstream APIs later require authentication, that risk would depend on user configuration rather than this tool’s stated design.

Although no fixed remote host is declared, its purpose is to generate MCP tools that interact with APIs from OpenAPI specifications, so user input or contextual data may be sent to external API endpoints defined in those specs; the actual egress targets depend on the loaded OpenAPI document.

The system flags this tool as executes-code, indicating that it runs locally as an MCP process/code component; this is a normal capability for this class of tool, but it still means there is local execution surface and the runtime should be constrained.

The materials do not state that it requires broad read/write access to local files, system resources, or persistent user data; based on the current description, data access appears mainly related to the provided OpenAPI specs and API invocation flow, with no clear sign of over-privileged access.

Positive factors are that it is open source under the MIT license and the code is auditable; however, it comes from a third-party registry, shows 0 stars, and has unknown maintenance status, which means external validation and maintenance signals are weak and supply-chain trust should be treated conservatively.