mcp-code-mode

The materials explicitly state that no keys or environment variables are required. No token requests, credential storage, or third-party account authorization are described, so the credential exposure surface appears low.

Neither the materials nor the objective checks list any remote endpoint, and there is no stated behavior of sending user data to external services. Based on the available information, no explicit network egress path is identified.

The objective checks indicate that this tool executes code, and the description explicitly says it runs Python code. This implies the ability to run local code/processes. Such capability is a normal high-privilege trait for an MCP tool and warrants caution around execution scope and runtime isolation, but by itself it does not justify a high-risk rating.

Although no README details are provided, local code execution typically can indirectly read or modify files and data accessible to the current runtime context. The materials do not indicate any extra privilege escalation, but they also do not define sandbox boundaries, so local data access should be treated with caution.

A positive factor is that there is an open-source repository, which improves auditability. However, the source is a third-party registry, the license is undeclared, community adoption is 0 stars, maintenance status is unknown, and README details are missing. Overall supply-chain transparency and maturity are limited, so source and dependencies should be reviewed before use.