io.github.n24q02m/wet-mcp

The materials state that API_KEYS and GITHUB_TOKEN are required, both of which are sensitive credentials; if over-scoped or exposed in logs/error output, they could be abused. No documentation is provided on intended use, least-privilege scope, or storage handling, so careful configuration is warranted.

The description includes web search, content extraction, and academic research, so outbound network access is inherent to its function. The materials do not specify concrete remote endpoints or data-flow details, which limits transparency, but they also do not show clearly suspicious or unrelated exfiltration hosts.

The system flags executes-code, indicating that the tool runs locally as an MCP service and executes code accordingly. This is a normal capability for MCP tools; the provided materials do not show requests for system privileges that clearly exceed its stated purpose.

Based on the description, the tool mainly handles user queries and fetched web, academic, and documentation content, and it will also read credentials from environment variables. Because the README is absent, it is unclear whether it reads/writes local files, where it caches data, or what retention policy it uses, so the data-access boundary is not fully transparent.

The source is the official registry and there is an auditable open-source repository with updates within the last year, all of which are meaningful risk-reducing factors. Although no license is declared and community stars are low, limiting community validation, there is currently no sign of closed-source distribution, abandonment, or obvious supply-chain red flags.