Small TypeScript starter with a local NodeBench MCP connection and verifiable development checks.
This is a boilerplate/template project, not a functional MCP server. The 16 tools are declared in scripts/verify-mcp.mjs but NO input schemas, parameter descriptions, or output schemas are visible in the source code. Tool definitions appear to be inferred from nodebench-mcp (a dependency), not explicitly defined here. The src/index.ts contains only a trivial hello() function. Without access to the actual tool registration code and schemas, per-tool scores are capped at 50 per the rubric. Most tools lack any visible description beyond their names. This is a starter template, not a production-ready server.
No input schemas visible for any tool. Tool definitions are inferred from nodebench-mcp dependency, not explicitly registered in this codebase. Cannot verify parameter types, constraints, or validation.
No tool descriptions visible in source code. LLMs cannot determine when or why to select each tool. Descriptions are required per pattern:tool-description.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 30 | 2026-07-28+ | v2 |
Tool naming inconsistency: 'getMethodology' uses camelCase instead of snake_case verb_noun convention. Should be 'get_methodology'. LLMs parse intent from names, inconsistent conventions reduce clarity.
Tool 'manipulate_screenshot' is vague. Does it crop, resize, annotate, or transform? Ambiguous names cause LLMs to guess wrong. Rename to specific action: 'crop_screenshot', 'annotate_screenshot', or 'resize_screenshot'.
Tool 'run_mandatory_flywheel' is unclear. 'Mandatory' and 'flywheel' are domain-specific jargon. Rename to reflect the actual action: 'execute_feedback_loop', 'run_iteration_cycle', or similar.