Loading ToolBench
Racing simulation QA database MCP server. Exposes the racesim.db data through Model Context Protocol tools, allowing LLMs to query batches, runs, metrics, and comparisons.
Compare metric means between candidate and baseline batches. Shows the delta (candidate - baseline) and percentage change for each metric.
Find all batches for a given scenario by name.
Get all runs and their metrics for a given batch.
Get aggregate statistics (mean, std, min, max) for each metric in a batch.
Get all metrics for a specific run.
Get threshold violations for a batch's telemetry data.
Get per-channel telemetry statistics for a batch (mean, std, min, max).
Get threshold definitions, optionally filtered by telemetry channel.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 12 | 2026-07-28+ | v2 |
Get car version details for all versions that have runs in a given scenario.
List all simulation batches with run counts and type (baseline/candidate).
List all car versions with their software/hardware configs and associated batches.
List all experiment scenarios with their parameters.
Query a specific metric across all batches, grouped by batch.