QAVE Studio Docs
QAVE Studio Docs
Sign inGet started — the Studio loop
Getting StartedFirst assurance runQuick assure
ConceptsLifecycleGlossaryHow QAVE Studio worksComing soon
Account & workspaceSignup and loginPeopleDemo creditsRolesSettings
Generate simulation suitesContext templatePII and field enhanceDimensionsAgents and plannerDraft review and Save
ExportExport formatsExport formats (detail)
Assurance SDKSDK overviewQAVE Studio suite fetch APIAgent and JSONL queueLangfuse bridge
Execute and tracesRecipe chatbot (reference SUT)TracingFetch tracing
AssureEvaluate APIRubrics and evidenceReading scores
History
WorkbenchGolden data
CoverageCoverage dashboardCoverage strategiesCoverage advisorTest case quality
RAG AssuranceRAG runsRAG datasetsRAG analytics
Project evalsError analysisUnderstanding scores
Studio guidesDimensionsTuplesConversation modesMulti-lingual testingWorkflowsHuman reviewBest practices
ReferenceAPI and packagesTroubleshooting
Assurance SDK

Agent and JSONL queue

Poll inputs.jsonl, call MCP evaluate_conversation, and record traces.

The SDK agent loads JSONL items and processes them in a loop.

Input shape

Each line is a JSON object, typically including id, trigger, mode, priority, messages, and criteria.

Run (local)

# From the qurator checkout used by your stack
PYTHONPATH=src python -m qurator.sdk.agent --poll-only

Configure chatbot / MCP URLs via environment (for example QURATOR_CHATBOT_ENDPOINT, QURATOR_CHATBOT_MCP_URL).

The reference recipe-chatbot does not yet expose MCP evaluate_conversation. For demos, point MCP at the mock SSE server. For real SUT traffic, call POST /api/chat or use the Langfuse path.

QAVE Studio suite fetch API

Project SDK keys and paginated /sdk routes for listing approved suites and cases.

Langfuse bridge

Fetch SUT traces from Langfuse, normalize, and submit to the assure Gateway.

On this page

Input shapeRun (local)