auraone

Connect agent systems, inspect tool traces, replay failures, compare candidates, and export regression evidence.

Author, validate, preview, calibrate, diff, and export criterion-level AI evaluation rubrics.

Synchronize robot episodes, inspect failures and recoveries, record review decisions, and export checksummed evidence.

Validate and render a disclosure record for judge prompts, calibration, bias, limitations, and intended use.

Import rubric JSON and response JSONL, validate them, run deterministic scoring, and inspect exact evidence in the browser.

Shared Proofline UI, runtime contracts, keychain, updater, and release infrastructure for AuraOne Open desktop Studios.

Validate, lint, diff, and convert portable AuraOne Rubric Schema v1 files.

Validate required Datasheet, Model Card, and Data Card sections and surface actionable review findings.

Measure inter-annotator agreement with modern metrics, uncertainty, ordinal support, and missing-data handling.

Probe judge behavior for position, verbosity, self-preference, paraphrase, anchoring, and calibration failures.

Scan MCP manifests, package metadata, source, and docs for tool, auth, filesystem, shell, network, and readiness risks.

Validate agent cards, capabilities, structured payloads, errors, cancellation, and task lifecycle behavior.

Compare before and after prompt and rubric directories and generate deterministic drift review notes.

Run executable rubric-spec v1 compatibility checks and emit reproducible conformance evidence.

Generate controlled synthetic annotator disagreement before real reviewer data is collected.

Browse, filter, validate, and reproduce public synthetic failure cases.

Integrate hosted AuraOne evaluation workflows through typed REST and GraphQL clients.

Provide repo-ready MCP, A2A, OTEL, replay, compare, and CI examples.

Validate and summarize human intervention and recovery segments in robot episodes.

Convert OpenTelemetry or Phoenix GenAI trace exports into local evaluation regression cases and manifests.