Loading ToolBench

Reading the current corpus…

mcs-mcp MCP server · F · ToolBench

mcs-mcp

MCP server for Jira-based workflow analytics, forecasting, and flow metrics (Monte Carlo simulations, cycle time analysis, WIP stability, aging analysis, CFD, throughput, yield, flow debt)

2026-07-28+STDIO0 tools

30/100

Definition Quality0

Protocol Readiness0

Supportability50

Spec posture

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

Static source inference · medium confidence

Current-spec patterns detected

Summary

Scoring was not performed

Issues (6)

CRITICALresponse-shaper ↗

Output schemas completely undocumented. No tool declares return type structure, field names, or types. LLMs cannot chain tools or extract data without guessing at response format.

analyze_cycle_timeanalyze_throughputanalyze_process_stabilityforecast_monte_carloanalyze_item_journeygenerate_cfd_data

HIGHconstrained-input ↗

Parameter constraints missing. 'bucketType', 'buckets', 'mode', 'percentile' lack enum definitions, min/max bounds, or format specifications. LLMs cannot infer valid values.

analyze_throughputanalyze_wip_stabilityanalyze_wip_age_stabilityanalyze_process_evolutionforecast_monte_carloforecast_backtest

HIGHtool-description ↗

Descriptions lack actionable LLM guidance. Descriptions are domain-specific (e.g., 'Understand baseline SLE') but do not explain WHEN to call the tool, WHAT to expect, or HOW to use results. Baseline is 194 chars; observed avg ~130 chars.

Recommendations

Document output schema for every tool. Example: 'Returns {cycles: [{issue_key, status_from, status_to, duration_days, started, completed}], p50: number, p85: number, p95: number}'. This enables LLM chaining and data extraction.
Add enum constraints to string parameters. Replace 'bucketType' description with: 'bucketType (enum): week|day|month|quarter' and validate on server side.
Expand descriptions to 150-250 chars. Include WHEN to call (e.g., 'Call after import_board_context to detect bottlenecks') and WHAT to do with results (e.g., 'Use p85 as SLE target').
Add parameter bounds. Example: 'buckets (integer): number of time buckets (1-52; default 12 for weekly analysis)'.
Add error recovery guidance. Example: 'If projectKey not found, call import_projects() to list available projects.'
Consolidate 'import_*' naming. Rename to 'discover_projects', 'discover_boards', 'load_board_context' to align with analytical verbs and reduce ambiguity.
Document forecast_monte_carlo parameter relationships. Clarify: 'Either specify newItemCount + targetDays (duration forecast) OR newItemCount alone (scope forecast). Mutually exclusive with customIssueList.'
Add per-tool dry-run capability for irreversible operations (e.g., cache expansions that may consume quota).

Score history

Overall score trend

↓ 19 points across a rubric change (v1 → v2)

30/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-22	F	30	2026-07-28+	v2
2026-03-09	F	49	-	v1