A gateway setup script and workflow examples for orchestrating multiple MCP tools without flooding AI context windows
This server has critical definition quality issues. Tool definitions are inferred from sample files (install.ps1, sample-workflow.js) rather than explicitly registered in source code. No actual MCP server implementation is visible, only installation scripts and workflow examples. The three tools (fetch, filesystem.write_file, postgres.query) lack proper schema validation, have minimal descriptions (10-30 chars), and no parameter type definitions beyond basic strings. Descriptions do not explain WHEN to use each tool, what prerequisites exist, or what the return structure is. No error handling guidance. No security considerations documented despite postgres.query accepting arbitrary SQL.
Fetch content from URLs
Write content to a file
Execute a PostgreSQL query
Tool definitions inferred from sample files, not explicitly registered in MCP server code. No actual server implementation visible.
Descriptions are under 20 characters and provide no context for LLM tool selection. 'Fetch content from URLs' does not explain when to use fetch vs other data sources, or what format is returned.
postgres.query accepts arbitrary SQL with no input validation, SQL injection protection, or permission gating. No documentation of required database credentials or connection setup.
Parameter schemas lack type definitions beyond 'string'. No enums, ranges, format constraints, or validation rules. 'params' array in postgres.query has no description of expected structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 28 | 2026-07-28+ | v2 |
No output schemas documented. LLMs cannot infer what fields to expect from fetch, write_file, or query results. No pagination, error structure, or response format guidance.