Enables AI agents to search service logs, group errors by signature, compare a candidate run against a baseline for regressions, and correlate which problems from other services appear before or after an error. All results are computed by the tools rather than the language model, so agents can only explain findings the code produced.
Enables AI agents and MCP clients to query trace, session, exception, prompt, dataset, score, and metric data for enhanced debugging and observability.
Enables AI agents to browse and filter experiment runs, retrieve metrics, logs, tables, and artifacts, plot comparisons as PNG images, and download full-fidelity training history for local analysis against a self-hosted Open Train server. It also provides diagnostic tooling that surfaces metric evidence and investigation leads without asserting causality.
Enables AI agents to access observability and evaluation data, including run history, span traces, LLM-as-judge evaluation results, and regression reports.
Enables coding agents to query OpenTelemetry traces, logs, metrics, and account usage from running services through read-only tools for trace lookup, verification, and usage inspection.