Loading ToolBench

Reading the current corpus…

neurameter MCP server · C · ToolBench

neurameter

github.com/kosukeuemura0220-del/neurameter

MCP server that exposes five cost-aware tools to any MCP-compatible AI agent for self-monitoring of cost, context, and budget

2026-07-28+STDIO5 tools

64/100

Definition Quality69

Protocol Readiness50

SupportabilityNot measured-

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

NeuraMeter MCP server has 5 tools with clear naming (verb_noun pattern) and reasonable descriptions (100-200 chars). However, parameter descriptions are sparse or missing entirely. Input schemas are present but lack depth: many parameters have no descriptions, and optional parameters are not clearly marked. Output schemas are not documented. Error handling guidance is absent. The server addresses a focused domain (cost/budget monitoring) but lacks the rigor expected for production agent tooling.

Tools (5)

neurameter_check_budgetread onlyauthsource verified72/100

Check whether the next operation fits within the budget. Returns budget limits, spending, remaining amount, and a decision (allow / warn / block).

neurameter_check_contextread onlyauthsource verified68/100

Analyse context-window efficiency for an agent. Returns utilization percentage, breakdown (system / conversation / toolResults), status, and optimization suggestions.

neurameter_get_cost_summaryread onlyauthsource verified72/100

Get a cost summary for the project or a specific agent. Returns total cost, budget remaining, top agents, and trend.

neurameter_get_recommendationsread onlyauthsource verified63/100

Get optimization recommendations based on historical data. Returns a list of prioritized actions with projected savings.

neurameter_log_optimizationwriteauthsource verified72/100

Log an optimization action performed by the agent (e.g. context summarization, model downgrade). Used for measuring optimization effectiveness.

Issues (5)

HIGHtool ↗

Output schemas not documented. Tools return data but LLMs cannot plan downstream calls or extract fields without trial-and-error.

neurameter_get_cost_summaryneurameter_check_contextneurameter_check_budgetneurameter_get_recommendationsneurameter_log_optimization

HIGHtool-description ↗

Parameter descriptions missing or incomplete. 'agentName' appears in 4 tools with identical description; 'optimizationType' lacks enum constraint or format guidance.

neurameter_get_cost_summaryneurameter_check_contextneurameter_check_budgetneurameter_log_optimization

HIGHrecovery-guide ↗

No error handling guidance. Tools do not document what happens on invalid input (e.g. invalid period, missing agent, budget exceeded). LLMs cannot self-correct.

neurameter_get_cost_summaryneurameter_check_contextneurameter_check_budgetneurameter_get_recommendationsneurameter_log_optimization

Recommendations

Document output schema for each tool. Example: neurameter_get_cost_summary returns {total_cost: number, budget_remaining: number, top_agents: [{name: string, cost: number}], trend: string}. This enables LLMs to extract and chain data.
Add enum constraint to 'period' parameter in neurameter_get_cost_summary and expand description: 'Time period for cost aggregation. Must be one of: today, week, month.'
Add descriptions to all parameters. For 'agentName': 'Optional. Filter results to a specific agent. If omitted, returns project-wide data. Agent names are case-sensitive.'
Add error guidance to tool descriptions. Example: 'Returns error if agentName does not exist. Call neurameter_get_recommendations with no agentName to discover available agents.'
Document result limits and pagination. Example: 'Returns top 10 agents by cost. Use offset/limit parameters (if supported) to paginate. Total count included in response.'

Score history

Overall score trend

First recorded score · v2 rubric

64/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	C	64	2026-07-28+	v2