Auraone Python Packages

auraone-sdk

Call hosted evaluation, analytics, training-export, and governance APIs from Python services, notebooks, and automation.

auraone-evalkit

Local-first AI evaluation and robotics review tools: EvalKit, Robotics ReviewKit, release evidence, and technical buying resources.

mcp-risk-linter

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

auraone-agent-studio-open

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

agent-trace-card

Turn one agent trace into portable Markdown, HTML, and JSON evidence cards.

a2a-contract-test

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

iaa-kit

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

rubric-spec

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

failure-gallery

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

datasheet-ci

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

contamination-audit

Scan evaluation data for n-gram overlap, canaries, answer patterns, hashes, and optional embedding similarity.

rubric-studio

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

eval-run-manifest

Create and validate a portable envelope for evaluation provenance, environment, artifacts, result digests, and optional signatures.

eval-adapter

Translate one rubric-spec rubric and run configuration into supported evaluation-framework inputs and exports.

eval-conformance-suite

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

embodiment-card

Validate and render structured cards for robot morphology, sensors, control, environment, data, and known limitations.

judge-bench

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

judge-card

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

lerobot-quality-gates

Check dataset structure, episodes, timestamps, media, actions, states, metadata, and release readiness.

robot-recovery-bench

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

otel-eval-bridge

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

synthetic-disagreement

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

prompt-rubric-drift

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

tool-call-replay

Normalize agent tool-call traces into deterministic replay artifacts and pytest cases.