0.2.3

What's Changed

• docs(examples): add CI-gate reference workflow (cherry-pick from fork#39) by @caohy1988 in #83
• fix(llm-judge): full prompt parity + execution_mode/fallback_reason by @caohy1988 in #85
• feat(evaluate): LLM-judge FAIL output with bounded feedback + --strict docs by @caohy1988 in #86
• fix(llm-judge): make AI.GENERATE template execute against current BigQuery by @caohy1988 in #87
• Bump version to 0.2.3 by @haiyuan-eng-google in #88

Full Changelog: 0.2.2...0.2.3

0.2.2

What's Changed

• feat(owl-import): SKOS support alongside OWL by @libei in #62
• feat(runtime): filter abstract upstream elements in both SDK adapters by @caohy1988 in #68
• docs: land in-repo implementation plan for concept index + runtime (issue #58) by @caohy1988 in #70
• docs: tighten scope and migration note on concept-index plan by @caohy1988 in #72
• Introduce RFC for entity resolution primitives by @haiyuan-eng-google in #73
• Fix evaluation logic for prebuilt gates and failure reporting by @haiyuan-eng-google in #78
• Fix evaluation logic for prebuilt gates and categorical metrics by @haiyuan-eng-google in #79
• docs: adopt Option 2 provenance columns for concept index (resolves #58 finding 1) by @caohy1988 in #80
• feat: Add agent improvement cycle demo by @evekhm in #61
• Update CodeEvaluator to compare raw observed values against budget by @haiyuan-eng-google in #81

Full Changelog: 0.2.1...0.2.2

0.2.1

What's Changed

• Enhance trace debugging with Span.tool_name and Trace.render options. by @haiyuan-eng-google in #64
• Remove blog draft after migration to dedicated blog repository by @haiyuan-eng-google in #65
• Fix text wrapper in Span.summary and Trace.final_response by @haiyuan-eng-google in #66
• Bump version from 0.2.0 to 0.2.1 by @haiyuan-eng-google in #67

Full Changelog: 0.2.0...0.2.1

Update to BigQuery Agent Analytics SDK to 0.2.0

What's Changed

• Implement V5 Context Graph with TTL import and temporal lineage by @haiyuan-eng-google in #13
• Update context graph design document by @haiyuan-eng-google in #14
• feat(ontology): add ontology models, loader, and tests by @libei in #9
• Add V5 context graph demo notebook by @haiyuan-eng-google in #15
• feat(ontology): add gm CLI with validate command by @libei in #11
• feat: batch_load materializer + deterministic V5 demo notebook by @haiyuan-eng-google in #22
• Update V5 Demo doc for better customer facing style by @haiyuan-eng-google in #23
• feat(ontology): add binding pydantic models by @libei in #16
• feat(cli): wire gm validate for binding files by @libei in #18
• Merge feat/compile into main (PRs #17–#19) by @libei in #24
• feat(cli): add gm compile command by @libei in #20
• docs(scaffold): add gm scaffold design doc by @libei in #25
• refactor: rename internal modules for clarity by @libei in #28
• docs: add ontology user manual by @libei in #29
• docs: move ontology user manual into package by @libei in #31
• feat(scaffold): add scaffold generator for table DDL and binding stubs by @libei in #33
• feat(scaffold): add gm scaffold CLI command and user guide by @libei in #34
• feat(owl-import): add OWL-to-ontology importer core logic by @libei in #35
• feat(owl-import): OWL-to-ontology importer with CLI and user guide by @libei in #36
• docs(user-manual): add CLI installation section by @libei in #37
• Implement ontology migration groundwork and DDL compiler bridge by @haiyuan-eng-google in #39
• fix: extract A2A_INTERACTION content in categorical evaluation transcripts by @evekhm in #42
• feat: Add Streamlit webapp for agent analytics metrics & visuals by @Saherpathan in #26
• Update project and dataset placeholders in notebook by @haiyuan-eng-google in #47
• Refactor and migrate to BigQuery Ontology Package with API cleanup and documentation by @haiyuan-eng-google in #50
• fix: strip markdown code fences from LLM classification responses by @evekhm in #46
• fix: increase max_output_tokens and retry NULL AI.GENERATE rows by @evekhm in #48
• fix: add ORDER BY before LIMIT for deterministic session ordering by @evekhm in #56
• Add quality evaluation report script by @evekhm in #54
• Implement telemetry labels for BigQuery jobs across SDK phases by @haiyuan-eng-google in #59
• Bump version from 0.1.0 to 0.2.0 by @haiyuan-eng-google in #60

New Contributors

• @evekhm made their first contribution in #42
• @Saherpathan made their first contribution in #26

Full Changelog: 0.1.0...0.2.0

0.1.0

What's Changed

• Add streaming evaluation, Dashboard V2, design docs, and CI workflow by @haiyuan-eng-google in #1
• Revise README for clarity and updated link by @haiyuan-eng-google in #3
• Overhaul README, add documentation indexes, and fix CI issues by @haiyuan-eng-google in #4
• docs: add ontology and context graph learning guide by @haiyuan-eng-google in #5
• Add Python 3.14 support and fix related test issues by @haiyuan-eng-google in #6
• Set version to 0.1.0 and implement PyPI release pipeline by @haiyuan-eng-google in #7
• Move ontology reference documentation to docs/ontology/ by @haiyuan-eng-google in #8

New Contributors

• @haiyuan-eng-google made their first contribution in #1

Full Changelog: https://github.com/GoogleCloudPlatform/BigQuery-Agent-Analytics-SDK/commits/0.1.0