2.4.0 (2026-07-07)

Features

• Add mTLS support for DiscoveryEngineSearchTool (8ba0e6a)
• Add mTLS support to Google API tools (3466586)
• add OpenAI Responses API support in labs (6b831d5), closes #3209
• Add support for reusing an existing SQLAlchemy AsyncEngine in DatabaseSessionService (f4c7e4c)
• add support for session TTL and expiration in Vertex AI session service (49d4441)
• adk: add Anthropic effort config handling, thinking parameter propagation, and conflict mitigation (4c862b9)
• agents: add ManagedAgent backed by the Managed Agents API (cf91b84)
• bigquery: expose thinking and tool-use token columns in analytics views (c14258d)
• bigtable: Support parameterized views with secure parameter inj… (14a24f2)
• HTTP traces for MCP server requests/errors (4c4f77a)
• Implement Workflow as Tool core feature (1263ed6)
• integrations: Add DaytonaEnvironment for remote sandbox workspaces (df6baf4)
• interactions: stream thought, media, code-exec and function-result deltas (b2dda6e)
• interactions: surface streamed grounding and final usage metadata (6a50b8d)
• memory: add Vertex AI load_profiles tool (fb2b3af)
• models: Add configuration options to Gemini class (037ec12), closes #3813
• models: surface and recover environment_id from interactions (81f9f2e)
• plugins: add otel correlation, custom_metadata allowlist, and column projection to BigQuery analytics (38d715c)
• scripts: add check for private-by-default new Python files (63561ce)
• support file_data URI references in GcsArtifactService (43115b2)
• support mTLS and GOOGLE_API_USE_MTLS_ENDPOINT for GDA client (e85a7b2)
• Support passing dynamic custom headers to LiteLLM via RunConfig (000d74d)
• Support provider-prefixed Gemini model IDs (816a87f)
• tools: add response_scheduling to control Live function response behavior (7443bfa)
• tools: exposed configurable parameter as property in McpToolset (cca8c56)
• tools: resolve built-in tools for managed-agent requests (f11d19d)
• tools: Support tuple tool parameters (a57c3e4)
• utils: Add support for nested state access in template injection (94c43a2)
• workflow: Allow ToolNode to accept JSON string or Content inputs (4e44632)

Bug Fixes

• adapt interactions conversion to google-genai 2.9 SDK changes (9f3aeef)
• Add credential_key to ApplicationIntegrationToolset and IntegrationConnectorTool (c6a94b8), closes #4553
• Add debug level logging to mcp sse agent sample (3962d0b)
• add DNS-rebinding protection to _OriginCheckMiddleware (9a4f479)
• add module blocklist for YAML agent config code references (6a5be34)
• add parameter validation for sync tools in ADK functions.py (9b9d27d)
• add trailing newline to check_new_py_files.sh (21512ea)
• Address mypy failures in google_api_tool/ (5301ffa)
• agents: persist output_key when before_agent_callback short-circuits LlmAgent (0e263f1)
• agents: preserve text streamed before tool calls in output_key (5a0b4af)
• agents: prevent path traversal in AgentTool config_path resolution (171ae9e)
• ApplicationIntegrationTool: implement dynamic mTLS endpoint resolution (37ca6fb)
• apply run_config custom_metadata to user event in node runtime (066fbce)
• artifacts: Preserve .text on GcsArtifactService load (ba51ff2), closes #3157
• auth: allow configuring OAuth prompt parameter (ac99770)
• auth: strip redirect_uri from credential_key (ffa1843)
• avoid mutating event validation input (6b385e4)
• Avoid overwriting existing isolation scope in stamp_event_branch_context (798207a)
• avoid yielding a None function-response event in live mode (4d88a52)
• catch RecursionError in safe JSON serialization helpers (5515800), closes #5411
• check if transfer target is a sibling agent (fa18d26)
• ci: Restrict GitHub Actions to main/v1 branches and main repository (8c4173e)
• cli: detailed error message on sse stream specifying stacktrack (error type on client side as well) (56b84e1)
• cli: enable interspersed option parsing for cloud_run command (7b049b9)
• cli: Ensure ADK compatibility in agent engine requirements (ea772b9), closes #5966
• cli: respect ignore files in adk deploy commands (ee79e71)
• cli: support flat-module agents in _determine_agent_language (7a553b7)
• concatenate list values in deep_merge_dicts during parallel tool call merge (fda2347), closes #5190
• concatenate list values in deep_merge_dicts during parallel tool call merge (1ff84eb), closes #5190
• Constrain artifact references to the caller scope (f863150), closes #6124 #6125
• correct misleading workflow-agent deprecation messages (53a8ab1)
• Derive sandbox reasoning engine from template or snapshot name (5afa9db)
• drop nonexistent log_query tool from session_state_agent sample (ade8577)
• emit standard OTel cloud.resource_id for Agent Engine telemetry (8fc25f1), closes #6247
• Enable history_config on Vertex AI now that it is supported (2920145)
• enforce agent-config args denylist when loaded under adk web (e506fa6)
• ensure AgentTool text output when skip_summarization is True (b983fcf), closes #3881
• exclude live HTTP clients from trace serialization (59868ea)
• exclude scripts/ from pre-commit compliance checks (c91fc75)
• exit connection cleanly on expected GoAway signal in bidi streaming (4d165ef)
• fingerprint cacheable context prefix (7c7f1e7)
• firestore: preserve update timestamps in list_sessions (99ba8ce)
• Fix branch parsing and unify branch stamping (9e3b43f)
• Fix Code Generation Template Injection via Filenames (e042b8d)
• Fix error swallowing in MCP session context (ccb8138)
• Fix event loop closed and thread leak errors in unit tests (2f2e03b)
• Fix instructions_utils matching invalid nested paths (20ba01c)
• Fix MCP debug client factory compatibility with keyword-only factories (1ad8864)
• fix polling loop in IAM Connector Credentials Provider when consent is pending (a614507)
• Fix regression in instructions_utils placeholder matching (9ecbaed)
• Fix Vertex AI Live API session replay on reconnect / modality switch (c007a87)
• guard against None converter results in RemoteA2aAgent (23ff66e)
• guard against None converter results in RemoteA2aAgent handlers (2e3d717)
• guard user_content assignment against None in EvaluationGenerator (6bc9c9f)
• Handle empty message in LiteLLM response (d3e793f)
• handle Pydantic models in safe_json_serialize for tool tracing (25d666a), closes #4629
• harden ContainerCodeExecutor sandbox by default (0a9ce0f)
• implement dynamic mtls endpoint resolution for parameter manager (62b9700)
• implement dynamic mtls endpoint resolution for secret manager (5d4a13f)
• import App lazily in _resolve_app to fix legacy Runner NameError (7481d07)
• in mcp mtls logic, check for authorization header case insensitively (3870032)
• LiteLLM Claude thinking blocks with display: "omitted" were lost (7461863)
• Litellm preserve streamed reasoning deltas in LiteLLM adapter (b9625bf)
• litellm: parse DeepSeek-V3 proprietary inline tool-call tokens (c5b2caa)
• LiteLlm: recognize assistant- prefix as valid OpenAI file ID (ea325c7), closes #5664
• LiteLlm: recognize assistant- prefix as valid OpenAI file ID (796964a)
• live: fix the return type hint of handle_function_calls_live (9865e2b)
• live: forward thinking config (71d8c84)
• live: keep streaming tool yields from completing turns (07aa1e0)
• Make EUC request args JSON-serializable (a181a39)
• map A2A Message.role to correct GenAI content role (fe08a9f)
• mcp: await async MCP header providers (c01e538), closes #6090
• mcp: propagate trace context in default streamable HTTP client (431e3c2), closes #4768
• migrate AgentRegistry base URLs from v1alpha to v1 (82432a3)
• models: gate Gemini cache creation on cacheable prefix tokens (1070036)
• N sized sliding window (d00ad67)
• preserve Anthropic thinking blocks and signatures in LiteLLM round-trip (febb250)
• preserve JSON-native types in A2A _serialize_value() (139c700), closes #5183
• prevent crash when parsing default values for typing.Any parameters (e360241)
• prevent empty responses after load_skill (7c79230)
• prevent VertexAiRagRetrieval from blocking the event loop (5898f37), closes #5033
• prevent warning logs when cancelling MCP session runner task (4aa0fd8)
• propagate Gemini grounding metadata from ModelResponse (c303c62)
• raise ValueError for unsupported MIME types in file_data URI path (42eceb3)
• reconnect on live connection 1011 error (f36d257), closes #5525
• recurse into JSON-native containers in A2A _serialize_value() (c08debc)
• refactor _redact_file_uri_for_log to dynamically extract file ID prefixes (file-, assistant-). This ensures compatibility with both current prefixes and future OpenAI/Azure file ID types. (ea325c7)
• remove unit test with wrong assumption about claude thinking parts (94fcdbb)
• remove unused userinfo_endpoint from GoogleApiToolset OIDC config (1509dcf)
• restore auth_token initialization for secret and parameter manager clients (46a2181)
• Rollback instruction util refactoring as its breaking internal customers (59fe9b3)
• sandbox nested persona template rendering in evaluation prompts (30493ba)
• scope single-turn node inputs to workflow branch (8d389e2)
• security: enable Jinja2 autoescape to prevent XSS in gepa sample (a721c1e)
• send correct field names for sandbox input files (2b7e08a)
• sessions: drop unsupported part_metadata before Vertex appendEvent (40a0279), closes #6014
• sessions: strip tzinfo for MariaDB in DatabaseSessionService (2f799d5)
• Set persist-credentials to false in release-update-adk-web workflow (79b8923)
• strip tzinfo from datetime for MySQL in create_session (a012bb7)
• surface error for empty STOP model turn in non-streaming mode (932a9b5), closes #5631
• telemetry: always export Agent Engine logs to stdout (e6df097)
• telemetry: emit gen_ai.workflow.nested for nested workflows (20197de)
• terminate infinite retry loop in RunSkillScriptTool on SCRIPT_NOT_FOUND (89d9bda), closes #5684
• tools: accept dict output_schema in SetModelResponseTool (#5469) (d69dedd)
• tools: handle tuple schemas in function parsing fallback (4fa7741)
• Truncate MCP http debug logs if greater than 1000 chars (3ebef82)
• Update custom gemini llm connection logic to be used for all 3_x models, not just 3.1 (8aff514)
• update litellm dependency constraint to >=1.84.0 (a912306)
• update mtls_utils import in secret_client.py (0b79f8d)
• use branch-scoped events for auth responses (f706a1e)
• use exchanged OAuth credential in ApplicationIntegrationToolset (e76df3d)
• use Modality enum for RunConfig.response_modalities (5c8c55a)
• use mTLS endpoint for Google OAuth2 token requests (ffe41f0)
• Use RELEASE_PAT for checkout in release-update-adk-web workflow (28b7721)
• validate local eval path segments (7b87f91)
• Wrap math formulas in math blocks in retry config guide (a40f199)

Performance Improvements

• avoid deepcopy of session contents when building LLM requests (400f512)
• create SQLite session schema once per service instance (10a2cc5)
• remove state injection when instruction has no placeholders (59970b6)
• skip event query in DB get_session when num_recent_events is 0 (7e8191a)

Documentation

• Add manual batch mode trigger for ADK Pull Request Triaging workflow (dbd4bb0)
• clarify is_final_response() is a public helper (978be4b), closes #4016
• fix typo in per-turn user simulator quality prompt (9a79fa1)
• index missing reference guides in adk-agent-builder SKILL.md (c922b08)
• remove trailing whitespace in dynamic_nodes guide (a7377b4)
• samples: Add ManagedAgent code-execution sample (527e3c1)
• samples: Add ManagedAgent sample using server-side google_search (969909f)
• Sort PRs by update time in list_untriaged_pull_requests (8679aa8)
• Streamline open_source_workspace/AGENTS.md (ea2ea73)
• Update ADK Pull Request Triaging Agent workflow to run periodically (b3f278a)
• update GoogleSearchTool docstring to refer to Gemini models (8b011f9), closes #12345
• Update sample-creator skill with agent topology graph rule (7423101)
• workflow: Add and update workflow developer guides (e5c7d20)

2.3.0 (2026-06-17)

Features

• add create_http_options to ContextCacheConfig for cache creation timeout (ffc9677), closes #4703
• add GCS first party toolset to ADK integrations (fb19e1a)
• Add log_level option for adk run CLI (1ac69a9)
• Add mTLS support in AgentRegistry client (03671c6)
• add request timeout to load_web_page (792775f)
• core: migrate core and CLI to enterprise parameters (68221f0)
• Create GEPARootAgentOptimizer (654145a)
• eval: expose user_simulator_config in generate_responses (e7a673c)
• gemma4: support Gemma4 in Gemini (573f043)
• integrations: Add E2BEnvironment for remote sandbox workspaces (92d608f)
• labs: add experimental Antigravity SDK agent wrapper (4cb27fd)
• lazily initialize the Vertex AI client in GCPSkillRegistry, and fix the import path for vertexai (65dca53)
• live: Handle input transcription differently for Gemini Live 3.1 models (048deea)
• live: support Live API translation config in RunConfig (463040f)
• Migrate McpToolset to AsyncAuthorizedSession for mTLS support (fe497a4)
• plugins: ADK 2.0 minimum producer cut for the BigQuery Agent Analytics plugin (e2676fc)
• report cached token counts for Anthropic and OpenAI models (b15c8a0)
• telemetry: support per-request OpenTelemetry configuration (abcaa08)
• Update gcp_auth client UI to support Remote Agents (57bdecf)
• utils: add GOOGLE_GENAI_USE_ENTERPRISE env var with deprecation fallback (4e85e9c)

Bug Fixes

• a2a: Preserve execution metadata in final events (e90b119)
• a2a: render HITL interrupt when prompt is in a data part (b9e7fca)
• a2a: set final=True for error TaskStatusUpdateEvent in event_converter (5efe53d)
• a2a: suppress part_metadata in Vertex AI mode (065f4ae)
• add a2a as a required dependency for agent_engine deployment (d2ebacb)
• adk: propagate exceptions from run_node in standalone mode (63841c3)
• adk: propagate isolation_scope to prevent history filtering loops (f39d75b)
• agents: await cancelled tasks in _merge_agent_run_pre_3_11 to prevent aclose() RuntimeError (9310ba7)
• api-registry to fetch all services (81b8067), closes #5478
• artifacts: Support nested API names (b99546b)
• auth: handle missing client-credentials scopes safely (a546bcf), closes #5345
• avoid UserWarning in _build_response_log when response has funct… (f022307)
• call to sanitize schema for complex union types (9808451)
• ci: add repository check to prevent workflows from running on forks (90bd38f)
• ci: Resolve missing sqlalchemy error in adk_release_analyzer (107dc38)
• cli: Serialize LiteLlm graph models safely (c1e852f)
• conformance: normalize tool declarations in replay verification (dd97e76)
• Default subagents to chat mode in build_node (ad560ce)
• deps: Require otel google-genai instrumentor >=0.7b1 for genai 2.x (2b8c80c)
• eval: handle failed inference results without invocations (9a6cf60)
• eval: handle unevaluated final response v2 results (5cfef01)
• eval: include function-call events in invocation_events when skip_summarization is set (5b16a86)
• eval: preserve custom eval metadata (780b0ab)
• Extract grounding_metadata from Live API server_content (8a294af)
• Fix silent dead end when conditional routes are unmatched (6af4562)
• Fix typing for create_client in mcp_session_manager (c6546a7)
• fix vertex_ai_session_service crashing when Agent Engine passes full resource names instead of short session IDs (60c55ad)
• flows: terminate invocation at tool-level EUC (883ff98)
• Format files to fix pre-commit failures (395848a)
• Format the files (9670ce2)
• gate pr-triage secrets on same-repository pull_request_target (0d20b7c)
• handle missing agent name in readonly context (a890399)
• improve error message when beautifulsoup4/lxml not installed for load_web_page (d9f189c), closes #4852
• live: history_config rejection on Vertex/Enterprise Live sessions (8f85260)
• live: propagate output token count in live API usage metadata (7e8965d)
• log diagnostics for empty or unparseable rubric auto-rater output (fe56f31), closes #5732
• make DatabaseSessionService visible in API docs (69ecf07), closes #4331
• Mock google.auth.default in test_fast_api.py (a7ceb3f)
• models: pass NOT_GIVEN to Anthropic when no system_instruction (3f505d2), closes #5318
• models: surface error when model returns STOP with empty content (ff95d2f)
• models: surface error when model returns STOP with empty content (423cd28)
• Move google-cloud-parametermanager to optional dependencies (0856093)
• Only send grounding_metadata for 3.1 live at the end of each turn (1f2e59b)
• otel: Handle empty contents in experimental semconv (d611f11)
• planners: allow BuiltInPlanner subclasses to override process_planning_response (f8e9195)
• planners: keep all leading parallel function calls (054da5d)
• plugins: write BigQuery analytics rows when invocation agent is None (bc08f46)
• preserve empty GCS text artifacts (8e2b06d)
• Preserve event details when output is delegated (a5a3f2e)
• preserve function call ids for litellm models (1ad348d), closes #2621
• prevent compaction from orphaning function responses (71b936b)
• prevent ReDoS in code block extraction (910e1c1)
• propagate model_version and other metadata in streaming responses (342b59d)
• remove developer notes from transfer_to_agent docstring (2a0b4e7)
• remove live event buffering in runner (4340208)
• remove the issue/PR analyze and fix agent workflows (9127feb)
• remove the issue/PR triage and fix agent workflows (66730e9)
• Reset retry attempt counter on successful connection (ca8baf1)
• restore GitHub-only changes dropped during v2 bring-over (cb48d01)
• sessions: Further fixes for DatabaseSessionService (f0ec997)
• sessions: honor zero recent events in database service (d9a672e)
• sessions: Prevent MissingGreenlet after append_event with asyncpg (06959b9)
• Set role='model' for request_input event Content (0c6974c)
• skills: enforce utf-8 encoding when materializing skill files on Windows (0cb4c81)
• skip crewai test on ImportError for pytest 9.1 compatibility (4aaf494)
• Stop interpolating release analyzer workflow inputs into shell commands (5a129a4)
• support non-Latin text in InMemoryMemoryService search (be1425b)
• surface MALFORMED_FUNCTION_CALL so on_model_error can recover (2fffcd9)
• tests: skip bash tool tests on Windows (9371f1b)
• tools: dereference draft-07 definitions in MCP tool schemas (c11ac7d)
• tools: handle missing 'request' key in AgentTool.run_async fallb… (8b09c48)
• update model to gemini-3.5-flash in session_state_agent sample (0aca7bf)
• use correct 'content' key in sandbox code executor input files (6262f94)
• utils: Preserve decorated type for @experimental and @working_in_progress (30d1910)
• utils: Preserve decorated type for @experimental and @working_in_progress (1ff0158)
• workflow: Preserve explicit single-turn contents (59f7bdf)
• workflow: Prevent replay divergence hang in sequence barrier (d88192c)
• workflow: Prevent Shared InvocationContext branch mutation (5c46937)
• workflow: Prevent silent drain of routed nodes in wait_for_output (ef8a5de)

Performance Improvements

• flows: skip async-rearrange when no function_responses (70b314b)
• test: Speed up unit test suite via parallelism and dedup (4e4bf84)

Code Refactoring

• Add diversion logic based on the auth provider resource name (d4ba521)
• Implement the auth provider using Agent Identity Credentials service (dc6fbd8)
• Move the IamConnectorCredential service depedency to a seperate file (c423fcd)
• otel: Add pure functions for constructing stable and experimental semconv logs (23c0826)
• Remove unused imports across src (b79096a)
• Separate PR analysis from triage for automation (10e5f07)
• telemetry: change agent and tool execution duration metrics from milliseconds to seconds (623c9bd)

Documentation

• add beginner explanation for single agent example (225fafc)
• add PyPI, Python version, downloads, and docs badges to README (04d278a)
• add unit guides for event.py, request_input.py and update adk-unit-guide skill (7d74a0a)
• add unit guides for task mode (f84a5b5)
• Align Python version to 3.10+ and update README badges (d3c21d7)
• clarify context cache min_tokens gating and 4096-token minimum (8c92cde)
• Fix ADK release analyzer session db saving error (991431f)
• fix formatting in multi-agent sample READMEs (fa82929)
• fix triaging agent sample typo (ef395c7)
• openapi: improve docs for session model (7a11b50)
• remove stale -b v2 flag from clone command in CONTRIBUTING.md (2e28e5d)
• skills: fix broken refs in adk-workflow skill (24a1b26)
• update llms.txt and remove build script (c66dc1d)
• Upgrade ADK release analyzer agent to use gemini-3.1-pro-preview model (d72bb7d)
• Add async and concurrency guidelines to ADK Style Guide (66e00db)

2.2.0 (2026-06-04)

⚠ BREAKING CHANGES

• agents: LlmAgent default model changed from gemini-2.5-flash to gemini-3-flash-preview (a preview model), ahead of the 2026-10-16 gemini-2.5-flash shutdown. Agents without an explicit model= now run on the new preview default; set model="gemini-2.5-flash" to retain prior behavior. DEFAULT_LIVE_MODEL is unchanged. (ad8b6c7)
• interactions: GenAI SDK v2.0.0 support renames the turn-based helpers in interactions_utils.py (e.g. convert_contents_to_turns → convert_contents_to_steps) and moves from "turns" to "steps" terminology. Update any direct callers. (da1d8f1)

Features

• Add --trigger_sources and ADK service options to cli_deploy_agent_engine (ffa057c)
• add AutoTracingPlugin for OpenTelemetry auto-instrumentation (bc3a4fa)
• add RubricBasedMultiTurnTrajectoryEvaluator (cae2337)
• agents: restore 1.x agent config wiring for backward compatibility (44cd116)
• api_server: Abort runs on client drops to avoid leaks (6a53357)
• BigQuery Agent Analytics reliability fixes (a5fa3da)
• distinguish input-required vs auth-required in A2A conversion (9d139ea)
• emit OTel gen_ai.client.* metrics natively (0bb329b)
• forward custom_metadata from run requests into the run config (460cb8c)
• include thoughts and tool calls in compaction summaries (bdb5582)
• interactions: update ADK to support Google GenAI SDK v2.0.0 (da1d8f1)
• models: Support turn_complete_reason in Live responses to capture safety info (9126acb)
• preserve A2A message metadata field in ADK event (d4d955d)
• raise explicit error for unsupported LiteLlm file attachments (8847f23), closes #5546
• sessions: add get_user_state(app_name, user_id) to BaseSessionService (d029bce)
• Support additional scopes and custom discovery doc in Google API Tools (dc6e293)
• tools: expose httpx_client_factory on RestApiTool and OpenAPIToolset (7eb9b3d)
• tools: Standardize request_input tool for proactive LLM clarification (afb0a64)

Bug Fixes

• a2a: Support to_a2a(Workflow) and reject non-agent root nodes (0478b02)
• accept Azure assistant file ids (b73679e), closes #5664
• add artifacts in each agent's .adk folder (bae5b1a)
• add future annotations import and prefix task models logger (2874874)
• add missing crop helper to data file helper lib (b5181cf), closes #4011
• add PEP 604 union syntax in function tool parameters (551445e)
• add telemetry metric assertions to the test's own agent (b7766ce)
• agents: restore abc.ABC base for BaseAgent and LlmAgent (020386a)
• allow internal builder assistant app name (f6e26cc)
• append trailing newline to runtime-config.json in ADK Web Server (4baccf6)
• auth: omit scope from OAuth2 token requests (6ce4b87)
• block path traversal in Agent Builder file tools (1fa7cda)
• dependencies: clarify missing Vertex AI extra (fde6a2b)
• deps: bump starlette and fastapi to address CVE-2026-48710 (81add39)
• eval: Support include_intermediate_responses_in_final in final_response_match_v2 (8519602), closes #5695
• Event.message honors subclass field (5bebfd4)
• exclude temp: state keys from Firestore session writes (a5db346)
• Fix path traversal in GCS skill extraction (Zip Slip) (2f15c6c)
• flows: preserve transparent config on live session reconnect (5ad1942), closes #5675
• Format the files to fix pre-commit failures (af8bfe0)
• guard peer agent mode access in agent transfer (bb16958), closes #5863
• live: Resolve 1007 error and support Gemini 3.1 Flash Live protocol (e5af12c)
• mcp: Prevent initialization hangs and task group leaks (334ef81), closes #5886
• migration: restrict unpickling of v0 actions blobs (9db48ce)
• models: Prevent grounding metadata loss in Gemini 3.1 (e896c62)
• parse noncanonical litellm tool call arguments (31cc5a1)
• populate user_content in resumed invocations (660bbd4)
• preserve media blocks in ollama content flattening (47ceeba), closes #4975
• runners: fall back to root agent when a resumed call author is not in the tree (a86efa6)
• runners: Preserve state_delta in NodeRunner path (c56bec8)
• sessions: guard None event.actions before reading state_delta (03ef3f6)
• Support generalized history config injection for Gemini 3.1 Live on Vertex AI (61a3933)
• terminate infinite retry loop in LoadSkillResourceTool on RESOURCE_NOT_FOUND (bc45ee6)
• tolerate context-likes without user_content or session in record_agent_invocation (0775da5)
• tools: add skill script dir to sys.path (9296198)
• tools: don't close parent's plugins from AgentTool's sub-Runner (2a68c4e)
• tools: Prevent broken skill tool references when prefix is set and support tool_filter (4366cca)
• tools: Shell escape path and range in ReadFileTool command (e16629b)
• validate session_id and enforce ownership in delete_session (b2916c7)
• workflow: Prevent incorrect chat agent wiring in graphs (d7aa7b5), closes #5868
• workflow: Resolve raw Content output crash on rehydration (4f992b0)

Performance Improvements

• flows: Resolve agent tool unions in parallel (ae95a97)

Code Refactoring

• agents: default model to gemini-3-flash-preview (ad8b6c7)
• tests: Consolidate event tests into test_event.py (77aeadf)
• tools: Split environment tools into single-class _tool files (1cc298e)
• update tool and agent retrieval functions to support asynchronous execution (e623b3b)

Documentation

• skills: Use default model in sample agent templates (2d465aa)

Maintenance (repo tooling, not library changes)

• skills: Add adk-issue skill to analyze and triage GitHub issues (be03166)
• skills: Add adk-review skill for rigorous change quality control (cc6f78c)
• skills: Automate PR triage and CLA verification (ce9011c)
• skills: Enforce PR assignment gates and stream metadata via stdout (4006fe4)
• skills: Split adk-issue skill to separate issue analysis from implementation (51b18eb)
• agents: Improve git hygiene in adk-pr-triage skill (62bcdd3)
• agents: Add issue closing support to PR triage skill (2748c1b)
• skills: Add rule to specify GitHub issues in commit messages (8f2c1e3)

2.1.0 (2026-05-23)

Features

• Add chart generation and artifact loading to data agent (db06416)
• Add support for creating sandboxes from templates and snapshots (cbd14eb)
• Add user.id to gen_ai.user.message log records for telemetry (eb379be)
• Fix error message telemetry for tool calls (e56c021)
• Preserve transcription event order in conversation trajectory (b3d0759)

Bug Fixes

• ci: Add python-dateutil dependency to stale-bot workflow (84fa984)
• ci: Prevent workflow failures in relocated adk_team samples (55cbc8c)
• ci: Use absolute path for PYTHONPATH in upload docs workflow (85223e6)
• cli: Fix --reload_agents for web (1307f8e)
• cli: Inform user to install optional dependency on missing google.cloud (57d677c)
• convert Union[Pydantic, Pydantic] tool args at runtime (104edc8), closes #5799
• Fix bug where grounding metadata in Gemini 3.1 live was being silently discarded (b9751eb)
• fix input and output transcription finished events for Gemini v3.1 (d17a2a3)
• lazy-import GCS evaluation managers in evals utility (5f91a9d)
• Make google-cloud-storage import lazy in skill utils (416775d)
• resolve circular import caused by llm_request (7e38fc8)
• Resolve circular import in base_tool (92cf192)
• tests: Append trailing newline to JSON test outputs (3329ced)
• tools: Prevent session drop on MCP tool error (933653c)
• update EditFileTool to handle cross-platform line breaks and escape regex characters (1f24553)

2.0.0 (2026-05-19)

ADK 2.0 General Availability

This release introduces v2.0.0 General Availability (GA) of the Google Agent Development Kit (ADK), establishing production-grade foundations for multi-agent workflows and advanced dynamic agent collaboration.

Core Architecture Highlights

Multi-Agent Workflow Engine

• Flexible Execution Graphs: Establishes a model-agnostic engine for orchestrating non-linear, conditional, and cyclical agent execution patterns.
• Intelligent Task Delegation: Introduces modular workflow abstractions enabling parallel sub-agent workers, nested hierarchical team structures, and resilient dynamic scheduling across complex task execution steps.

Dynamic Agent Collaboration

• Native Inter-Agent Routing: Provides seamless orchestration for inter-agent messaging, control state handoffs, and context variable propagation across collaborative multi-agent flows.

2.0.0-beta.1 (2026-04-21)

Highlights

• Transition to Beta: Updated documentation to reflect the project's move to Beta phase (84e792fd).
• Security Fix: Resolved a potential RCE vulnerability related to nested YAML configurations (2b54c4ac).
• Documentation & Style: Modularized the ADK style guide and added new topics (9bb21795).
• Workflow Orchestration Core:
  • Added the full Workflow(BaseNode) graph orchestration implementation (c25d86f1).
  • Introduced NodeRunner for per-node execution isolation (0b3e7043).
  • Added DefaultNodeScheduler for standalone node resume via ctx.run_node() (a68e70d5).
• Agent Execution Models:
  • Added explicit ReAct loop nodes to replace legacy single agent flows (94453619).
  • Supported BaseNode as the root in both the Runner and the CLI (91b56b9a, 1c2a81bb).
• State & Resumption:
  • Added Human-in-the-loop (HITL) resume via event reconstruction (ca327329).
  • Added lazy scan deduplication and resume for dynamic nodes (d600b195).
  • Supported flushing state/artifact deltas onto yielded events (00153088).
• Performance:
  • Optimized execution by bypassing the Mesh for leaf single-turn LlmAgent instances (d864917f).

Features

• Workflow Orchestration: Added Workflow(BaseNode) graph orchestration implementation, support for lazy scan deduplication and resume for dynamic nodes, partial resume for nested workflows, and state/artifact delta bundling onto yielded events.
• CLI and Web UI: Added support for Workflow graph visualization in web UI, improved graph readability with distinct icons and shapes, and active node rendering in event graph.
• Documentation: Added reference documentation for development (skills like adk-style, adk-git, and observability architecture).

2.0.0-alpha.2 (2026-03-26)

Bug Fixes

• add agent name validation to prevent arbitrary module imports (be094f7)
• add protection for arbitrary module imports (0019801), closes #4947
• Default to ClusterIP so GKE deployment isn't publicly exposed by default (27aed23)
• enforce allowed file extensions for GET requests in the builder API (68ece4e)
• Exclude compromised LiteLLM versions from dependencies pin to 1.82.6 (79ed953)
• gate builder endpoints behind web flag (584283e)
• Update eval extras to Vertex SDK package version with constrained LiteLLM upperbound (a86d0aa)

Features

Introduces two major capabilities:

• Workflow runtime: graph-based execution engine for composing deterministic execution flows for agentic apps, with support for routing, fan-out/fan-in, loops, retry, state management, dynamic nodes, human-in-the-loop, and nested workflows
• Task API: structured agent-to-agent delegation with multi-turn task mode, single-turn controlled output, mixed delegation patterns, human-in-the-loop, and task agents as workflow nodes

1.39.1 (2026-08-27)

Bug Fixes

• a2a: Port event action allowlist, agent card description source and resume flattening to v1 (#6801) (1f898a6)
• artifacts: Port app-scoped file artifact storage and layout-derived payload paths to v1 (#6795) (0ae2bff)
• artifacts: Port artifact reference scoping and path-segment validation to v1 (#6794) (efc5bb8)
• cli: check the Host header on every local API server request (v1) (#6789) (15bf230)
• cli: Port deploy ignore-file handling to v1 (#6793) (8467bf0)
• close the residual gaps from the v1 backport review (#6890) (218061e)
• integrations: send api registry credentials only to google api endpoints (v1) (#6889) (1e9e117)
• mcp: Port config stdio opt-in and reserved tool name checks to v1 (#6796) (28b32b4)
• models: Port Gemini context cache fingerprint scoping to v1 (#6804) (f2f8456)
• plugins: sanitize BigQuery analytics rows and build GCS paths per call (v1) (#6808) (80e56de)
• Port A2A agent card RPC target origin validation to v1 (#6800) (ccb4812)
• Port agent loader and file path resolution fixes to v1 (#6788) (8bc927f)
• Port agent-config code-reference and config_path validation to v1 (#6790) (15ae713)
• Port branch-scoped auth responses and OAuth2 nonce forwarding to v1 (#6803) (5817d30)
• Port code executor container and pod isolation defaults to v1 (#6807) (1b4da44)
• Port credential and inline-data field exclusions in logs, reprs and spans to v1 (#6797) (7c2075c)
• Port eval identifier validation and persona template rendering to v1 (#6806) (0a4a5c6)
• Port migration unpickle allowlist and database URL redaction to v1 (#6791) (6deda28)
• Port session id validation, thread id scoping and Vertex RAG cleanup fixes to v1 (#6809) (3fec8c5)
• Port skill resource path and skill name validation fixes to v1 (#6805) (1353e8d)
• Port URL validation, client timeout and credential scoping fixes to v1 (#6802) (7b6547f)
• Route thread-pool and live-mode tool dispatch through run_async (v1) (#6799) (48c56c3)
• Scope custom eval metrics per registry and resolve their function path from the eval config (v1) (#6792) (9886410)
• sessions: accept a full session resource name again on v1 (#6888) (ef4f1e0)
• Validate user message parts and agent transfer targets (v1) (#6798) (d830355)

1.39.0 (2026-08-17)

Features

• live: use RunConfig.session_resumption.handle when opening a live session (#6770) (090869b)
• support audio_stream_end for realtime input (#6699) (5e3b997)

Bug Fixes

• live: stop background tool tasks when a live agent run ends (#6769) (5fb5ff1)
• validate session initialization events (v1) (#6771) (e7dba7a)

Features

• live: forward safety_settings from generate_content_config to the Live API (#6641) (99c1aa3)

Bug Fixes

• avoid shell execution in ReadFileTool ranged reads and keep local environment file access in workspace (#6597) (c680728)
• deps: exclude nltk 3.10.1, which breaks venvs living inside the working directory (v1) (#6607) (f25bb90)
• eval: handle failed inference results without invocations (cherry-pick) (#6565) (4555273)
• live: transfer to the target agent regardless of function response order (#6608) (302885b)
• Port tool confirmation security and re-validation fixes to v1 (#6575) (4783364)

1.37.0 (2026-07-30)

Features

• live: Support asynchronous non-blocking tools (#6522) (c299420)

1.36.2 (2026-07-21)

Bug Fixes

• Update custom gemini llm connection logic to be used for all 3_x models, not just 3.1 (#6438) (4ef72fe)

1.36.1 (2026-07-06)

Bug Fixes

• exit connection cleanly on expected GoAway signal in bidi streaming (v1) (#6264) (de2abdf)
• Fix Vertex AI Live API session replay on reconnect / modality switch (#6218) (4caf782)

1.36.0 (2026-06-22)

Features

• interactions: support for GenAI SDK upgraded to 2.9+ (#6191) (7a9152a)

1.35.2 (2026-06-17)

Bug Fixes

• remove live event buffering in runner (#6151) (afe4083)

1.35.1 (2026-06-15)

Bug Fixes

• Only send grounding_metadata for 3.1 live at the end of each turn (#6129) (847a259)
• streaming: Ensure final partial=False frame is always yielded (#6096) (6e59c61)

1.35.0 (2026-06-10)

Features

• live: Handle input transcription differently for Gemini Live 3.1 models (#6045) (ecfdaf5)

Bug Fixes

• add missing Gemini imports in base_llm_flow (#5943) (6d027b4)
• flows: Reset reconnect attempts on connection success (#6042) (87abf23)
• models: Default grounding metadata for Gemini 3.1 live (#6018) (fafafb3)
• Support generalized history config injection for Gemini 3.1 Live on Vertex AI (#5999) (aafd97f)

1.34.2 (2026-06-01)

Bug Fixes

• Fix bug where grounding metadata in Gemini 3.1 live was being silently discarded (9b6b9e9)
• fix input and output transcription finished events for Gemini v3.1 (13763d7)
• tools: Prevent session drop on MCP tool error (1fd406b)

1.34.1 (2026-05-22)

Bug Fixes

• Fix bug where grounding metadata in Gemini 3.1 live was being silently discarded (9b6b9e9)
• fix input and output transcription finished events for Gemini v3.1 (13763d7)
• tools: Prevent session drop on MCP tool error (1fd406b)

1.34.0 (2026-05-18)

Features

• a2a: add support for persistent task stores (cd78d87)
• add general support for Gemini Live API in ADK evaluate (790c9be)
• Add mTLS support to Google Cloud Telemetry exporter (cfe8d2c)
• add support for A2aAgentExecutor factory in to_a2a() function (115124c)
• add support for non-ADK produced input-required events (6e53472)
• Added config option to include tool calls/responses in conversation history passed to user simulator (baf7efb)
• ci: add Gemini auto review and invoke workflows (fd8b492)
• Implement GCPSkillRegistry in ADK (88ebd42)
• Implement Skill Registry in ADK (380d261)
• Make Agent Skill description validation more informative (9f38973)
• Simplify data retrieved handling of ask_data_agent tool and ask_data_insights tool (48f1b30)
• Support OAuth PKCE in McpToolset (e7316dc)

Bug Fixes

• agents: fix visibility of output_key state delta in callbacks (0524797)
• anthropic: map negative thinking_budget to adaptive thinking (03b915b)
• auth: persist refreshed OAuth2 credentials to store (218ea76), closes #5329
• auth: remove unneeded OAuth flows (c35a579)
• avoid pre-serializing dict values in Interactions API to prevent double-escaping (85f397d)
• cache: enforce CacheMetadata active-state invariant (76b9f0b)
• cache: handle fingerprint-only metadata in performance analyzer (9c5de58)
• Catch OSError when importing AnthropicLlm (91cb5c6)
• evaluation: handle none config in per_turn_user_simulator_quality (eed9bd3), closes #5677
• fallback to project id if crendetials don't contain quota project (e377cb5)
• Fix missing dynamically loaded tools in SkillToolset during the same invocation (f9097cb)
• live: ensure sub live agent doesn't inherit session resumption handle from parent live agent to avoid interrupting the conversation (8dd9147)
• models: preserve string content in Anthropic tool_result blocks (9a1e75f), closes #5358
• models: preserve tool_use IDs for Anthropic models on session resume (327c45f), closes #5074
• models: treat empty GenerateContentResponse without prompt feedback as successful (0cb9ae9)
• only serialize llm_response to json if it will be included in the trace (1284493)
• Preserve live_session_id in function call handling (07a9a01)
• Prevent compaction of events involved in Human-in-the-Loop interactions (bb2efb6)
• raise eagerly on importing AgentRegistry if a2a-sdk is missing (33cf6cb)
• small: Convert events to the A2A format while respecting user vs agent role (59f7347)
• tools: preserve code_execution_result and executable_code in AgentTool (7e61b51), closes #5481
• tools: Prevent AnyIO CancelScope task boundary violations during MCP session creation failure (4309159)
• Update model name in hello_world agent (192f19d)
• Update model to gemini-3-flash-preview in hello word agent sample (6d89d21)
• Update model to gemini-3-flash-preview in session state agent sample (2d423e8)
• use tool_responses role for gemma4 models in LiteLLM integration (3d07960), closes #5650

Performance Improvements

• lazy-load service registries and split apps.app to cut cold start ~8% (bd062ec)
• models: guard debug log evaluation with isEnabledFor (57d8fc7)
• utils: cache find_context_parameter introspection (ec54bd4)

Code Refactoring

• Make the "a2a_metadata" string a constant that can be depended on by extension developers (0821f2d)

1.33.0 (2026-05-08)

Features

• add BufferableSessionService (0bc767e)
• apigee: allow injecting credentials into ApigeeLlm (ce578ff)
• Make ADK environment tools truncation limit configurable (83ae405)
• models: add get_function_calls and get_function_responses to LlmResponse (22fae7e)

Bug Fixes

• catch genai.ClientError when sandbox is missing (69fa777), closes #5480
• double append bug (f8b4c59)
• Filter out video events with inline data from being stored in session (88421f8)
• fix fork detection, correct offload limits, and add response logging in BigQuery plugin (9d1bb4b)
• hot reload agents for adk web (740557c)
• Only append skills to system instruction if ListSkillsTool isn't available (01f1fc9)
• prevent state_delta overwrite on function_response-only events (fc27203, 211e2ce), closes #3178
• Raise a clear actionable error when CustomAuthScheme lacks a registered AuthProvider (83f9817)
• should use app_name instead of req.app_name (8286066)
• simulation: Add error message when LlmBackedUserSimulator returns empty response (fb92aad)
• Update expressmode api call to include default api key param (e833995)
• use asyncio.sleep to avoid blocking event loop (3a1eadc)
• Use project and location instead of API key when deploying to agent engine (398f28f)

Code Refactoring

• adjust computation of workflow.steps metric and add new unit tests (03d6208)
• remove input.type and output.type attributes from adk metrics (9559968)

1.32.0 (2026-04-30)

Features

• Add an option to prevent the SaveFilesAsArtifactsPlugin from attaching reference file parts to the message (987c809)
• add credentials parameter to BigQueryAgentAnalyticsPlugin (34713fb)
• Add express mode onboarding support to adk deploy cli (2b04996)
• add native OpenTelemetry agentic metrics (6942aac)
• Add OpenTelemetry tracing for event compaction (c65dd55)
• Add sample agent demonstrating 2LO, 3LO, and API Key auth via GcpAuthProvider (909a8c2)
• Add support for Anthropic's thinking blocks (16952bd)
• Add support for excluding predefined functions in ComputerUseToolset (d760037)
• Add support for refusal messages in ApigeeLlm (d6594a1)
• Added indication of user message in history event list (662354a)
• Allow user to define credential_key for McpToolset (282db87), closes #5103
• analytics: add support for logging LLM cache metadata to BigQuery (02deeb9)
• eval: add evaluate_full_response option to rubric-based evaluation (#5316) (7623ff1)
• live: Add save_live_blob query parameter to /run_live endpoint (36ab8f1)
• mcp: gracefully handle tool execution errors and transport crashes (7744cfe)

Bug Fixes

• accumulate list values when merging parallel tool call state_delta (b0b8b31), closes #5190
• Add support for overriding the API version in GoogleLLM (1cdd1e7)
• auth: isolate resolved credentials in context to prevent race conditions and data leakage (5578772)
• avoid double-execution of sync FunctionTools returning None (78a8851), closes #5284
• block RCE vulnerability via nested YAML configurations in ADK (74f235b)
• bump Vertex SDK version (6380f6a)
• cancel siblings in parallel function calling on failure (49985c9)
• Capture and include LLM usage metadata in summarized events (5ce33b9), closes #4014
• catch ValueError in safe-JSON serializers for circular refs (70a7add), closes #5412
• deps: bump litellm cap to >=1.83.7 to admit CVE patches (6d2ada8)
• Disable bound token for mcp_tool (4c0c6db)
• fix dataset location handling in BigQueryAgentAnalyticsPlugin (c263426)
• Fix exception handling and argument order in ReflectRetryToolPlugin (1deab6d)
• Fix GcpAuthProvider to return capitalized Bearer scheme (ad937fe)
• fix lifecycle issues with credentials in BigQuery Agent Analytics Plugin (a69f861)
• Fix malformated skill.md (9a0d2f7)
• Fix misplaced pytest decorator on helper dataclass in 2LO integration tests (2343973)
• Fix RecursionError in ADK framework by adding circular reference detection to schema resolution (7de5bc5)
• fix rewind to preserve initial session state (af1b00a), closes #4933
• Fix SSRF and local-file access in load_web_page (0447e93)
• handle None state values in skill_toolset after session rewind (a977aa3)
• litellm: emit input_audio for audio inline_data parts (4073238), closes #5406
• live: mark all agents' Event as from other agents (48b7a64)
• live: treat input transcription as user message (ae1f2e6)
• optimization: handle None metric scores in LocalEvalSampler (#5415) (684a6e7)
• otel: change gen_ai.tool_definitions to gen_ai.tool.definitions (029b87d)
• preserve cache fingerprint stability on creation failure (4d5438c)
• preserve empty-string text parts in A2A converter (2d61cb6)
• preserve function call IDs for Anthropic models (f0c787f)
• Prevent LoopAgent from resetting sub-agent state on pause (8846be5)
• Quote user_id literals in VertexAiSessionService list filters (bdece00)
• read_file/write_file path type mismatch in BaseEnvironment and LocalEnvironment (782796f)
• relax EventActions.state_delta value type to Any (dbec8e9)
• remove exclude_unset=True to correctly serialize pydantic types (f95ac48)
• samples: Upgrade google-adk to 1.28.1 to fix vulnerability (b848390)
• Sanitize user_id derived from PubSub subscription and Eventarc source (0c4f157), closes #5324
• Scope Vertex RAG memory display names (784350d)
• Use correct camelCase functionCallId (c87ee1e)
• web oauth flow and trace view (87cd310)
• yield tool_call_parts immediately in live mode to unblock Gemini 3.1 tool calls (f57b05d)

Performance Improvements

• lazy-load optional providers and auth chain to cut cold start ~25% (66bfedc)

Code Refactoring

• move exception handling from metric emission into instrumentation handlers (62d7ee0)
• tests: Refactor tests to explicitly handle JSON_SCHEMA_FOR_FUNC_DECL feature flag (b580891)
• Use artifact_service.load_artifact during rewind (c3d50db), closes #4932

Documentation

• gemini: show subclass pattern for custom Client config (34c7505), closes #3628
• update output_schema docstring to reflect support for tools and output_schema together (e1e652d)
• Update README with instructions for installing ADK extensions (f2a1179)
• use sphinx-click to generate docs for google.adk.cli (f455974)

1.32.0 (2026-04-21)

Bug Fixes

• block RCE vulnerability via nested YAML configurations in ADK (e283ea0)
• bump Vertex SDK version (21ebe29)
• Disable bound token for mcp_tool (d0afddd)
• web oauth flow and trace view (c3dafdb)

1.31.0 (2026-04-16)

Features

• New Live UI: Redesigned the real-time chat and testing interface for a cleaner, more responsive user experience.
• Session Display Names: Added the ability to name and organize debug sessions for easier tracking.
• New Events and Trace View: Overhauled the execution log into a structured trace for better visibility into agent logic.
• New Graph View: Introduced a visual canvas to map out and manage agent architectures and node relationships.
• Event Filtering: Added granular controls to filter the event stream by message type, tools, or errors.
• New Computer Use View: Implemented a dedicated UI for visualizing screenshots and logs during computer-use tasks.
• Add "google-adk" user agent to Parameter Manager and Secret Manager clients (b8e8f6b)
• Add support for memories.ingest_events in VertexAiMemoryBankService (d69477f)
• Add Vertex AI Agent Engine Sandbox integration for computer use (7686848)
• Firestore support (1a9df8f)
• live: Add live_session_id to LlmResponse (bf84e2c)

Bug Fixes

• Bump minimum mcp version from 1.23.0 to 1.24.0 (494c360)
• cli: correct console URL path after adk deploy agent_engine (64ed1a6), closes #5336
• execute on_event_callback before append_event to persist plugin modifications (454188d), closes #3990
• make _EvalMetricResultWithInvocation.expected_invocation Optional for conversation_scenario support (#5215) (a4c9387)
• Pass in auth headers with header provider instead of connection params (e12b0af)
• populate required fields in FunctionDeclaration json_schema fallback (9b9faa4)
• Resolve BigQuery plugin issues with A2A transfers, spans, and metadata (9ca8c38)
• upgrade google-genai lower bound (8bc5728)

Code Refactoring

• live: Use send_client_content to send conversation history (67dc2eb)
• live: Use send_tool_response for function responses (70c5fc8)

Documentation

• update MCP Toolbox branding, binary version, and asset references (47fa7b7)

1.30.0 (2026-04-13)

Features

• Add Auth Provider support to agent registry (f2c68eb)
• Add Parameter Manager integration to ADK (b0715d7)
• Add support for Gemma 4 models in ADK (9d4ecbe), closes #5156
• allow users to include artifacts from artifact_service in A2A events using provided interceptor (e63d991)
• emit a TaskStatusUpdateEvent for ADK events with no output parts but with event.actions (dcc485b)
• Live avatar support in ADK (a64a8e4)
• live: expose live_session_resumption_update as Event in BaseLlmFlow (2626ad7), closes #4357
• Promote BigQuery tools to Stable (abcf14c)
• samples: add sample for skill activation via environment tools (2cbb523)

Bug Fixes

• Add "gcloud config unset project" command to express mode flow (e7d8160)
• avoid load all agents in adk web server (cb4dd42)
• Change express mode user flow so it's more clear that an express mode project is being created (0fedb3b)
• Custom pickling in McpToolset to exclude unpicklable objects like errlog (d62558c)
• Fix credential leakage vulnerability in Agent Registry (e3567a6)
• Include a link to the deployed agent (547766a)
• preserve interaction ids for interactions SSE tool calls (9a19304), closes #5169
• validate user_id and session_id against path traversal (cbcb5e6), closes #5110

Changelog

1.29.0 (2026-04-09)

Features

• Add auth scheme/credential support to MCP toolsets in Agent Registry (7913a3b)
• add ability to block shell metacharacters in BashTool (23bd95b)
• add configurable resource limits for subprocesses in BashTool (1b05842)
• Add configurable view_prefix to BigQueryLoggerConfig (37973da)
• Add custom session id functionality to vertex ai session service (e1913a6)
• Add Description column to SKILL.md and update terminology (435f7c7)
• Add Easy GCP support to ADK CLI (8850916)
• Add regional endpoint support to SecretManagerClient (19ac679)
• Add support for model endpoints in Agent Registry (eb4674b)
• auth: Add public api to register custom auth provider with credential manager (a220910)
• auth: Pass consent_nonce to Agent Frontend (9fec503)
• auth: Support additional HTTP headers in MCP tools (b3e9962)
• bigquery: Add ADK 1P Skills for ADK BQ Toolset (4030c0d)
• environment: Add EnvironmentToolset for file I/O and command execution (9082b9e)
• environment: Add LocalEnvironment for executing commands and file I/O locally (f973673)
• Implement robust process group management and timeouts in BashTool (f641b1a)
• live: Added in 1.28.1, support live for gemini-3.1-flash-live-preview model (8082893)
• Option to use shallow-copy for session in InMemorySessionService (16a1a18)
• Propagate context to thread pools (83393ab)
• refresh credentials if token is missing in the common code and samples (1445ad5)
• Remove use of raw_event field in vertex ai session service (642d337)
• skill: Standardize skill tools and make script arguments flexible (9e73ab8)
• Support AgentRegistry association (6754760)
• Support loading agents from Visual Builder with BigQuery-powered logging (2074889)
• Support propagating grounding metadata from AgentTool (d689a04)
• Support short options and positional arguments in RunSkillScriptTool (2b49163)
• Use raw_event field in vertex ai session service for append and list events (6ee0362)
• Use raw_event to store event data in vertex ai session service (9da9dee)

Bug Fixes

• Add A2ATransport.http_json to the default supported transports list (7dd9359)
• add httpx_client_factory support to SseConnectionParams (815ebb4)
• adk: redact credentials in BigQuery analytics plugin (a27ce47)
• api client initialization logic to be mutually exclusive between ExpressMode and GCP projects (4ffe8fb)
• avoid load all agents in adk web server (ede8a56)
• Cache BaseToolset.get_tools() for calls within the same invocation (92cad99)
• cli: fail Agent Engine deploy when config file path is invalid (bbad9ec)
• Disable tool caching for skill toolset (064f0d2)
• Disallow args on /builder and Add warning about Web UI usage to CLI help (dcee290)
• empty events_iterator assignment (898c4e5)
• environment: fix package references (add8e86)
• Fix RemoteA2AAgent deepcopy errors (6f29775)
• Fixes for initializing RemoteA2aAgent - passing in preferred transport, protocol version, and auth headers (0f3850f)
• Generate IDs for FunctionCalls when processing streaming LLM responses (fe41817), closes #4609
• Handle merging lists in deep_merge_dicts (cdb3ff4)
• In memory session service to evaluate dictionary keys and value into an isolated snapshot sequence before starting loop (f75de59)
• include intermediate subagent final response events in evaluation intermediate data (f8a6bd7)
• live: Handle live session resumption and GoAway signal (6b1600f), closes #4996
• move BigQueryAgentAnalyticsPlugin import inside get_runner_async (6fd0f85)
• Safer fix for UI widget merging in ADK (0e71985)
• Small fixes for express mode (3a374ce)
• sync callbacks with call_llm span (b2daf83)
• tools: handle toolset errors gracefully in canonical_tools (5df03f1), closes #3341
• truncate error_message in v0 schema to prevent VARCHAR overflow (62daf4f), closes #4993
• update toolbox-adk and toolbox server versions (1486925)

Code Refactoring

• Move SecretManagerClient to google.adk.integrations.secret_manager package (1104523)
• Remove the session events dependency from A2aAgentExecutor (aaa03ac)

Documentation

• adk: clean up remote triggers README to remove internal references (ccac461)
• Update the MCP Toolbox docsite with the new URL (a60baca)

1.28.1 (2026-04-02)

Features

• live: support live for gemini-3.1-flash-live-preview model (ee69661)

Bug Fixes

• Disallow args on /builder and Add warning about Web UI usage to CLI help (f037f68)
• live: Buffer tool calls and emit them together upon turn completion (081adbd)

1.28.0 (2026-03-26)

Features

• a2a: add lifespan parameter to to_a2a() (0f4c807), closes #4701
• Add a new extension for the new version of ADK-A2A integration (6f0dcb3)
• Add ability to run individual unit tests to unittests.sh (b3fcd8a)
• Add database_role property to SpannerToolSettings and use it in execute_sql to support fine grained access controls (360e0f7)
• Add index to events table and update dependencies (3153e6d), closes #4827
• Add MultiTurn Task success metric (9a75c06)
• Add MultiTurn Task trajectory and tool trajectory metrics (38bfb44)
• Add slack integration to ADK (6909a16)
• Add Spanner Admin Toolset (28618a8)
• Add SSE streaming support to conformance tests (c910961)
• Add support for Anthropic's thinking_blocks format in LiteLLM integration (fc45fa6), closes #4801
• Add support for timeout to UnsafeLocalCodeExecutor (71d26ef)
• auth: Integrate GCP IAM Connectors (Noop implementation) (78e5a90)
• bigquery: Migrate 1P BQ Toolset (08be442) (7aa1f52) (d112131) (166ff99)
• enable suppressing A2A experimental warnings (fdc2b43)
• Enhance AgentEngineSandboxCodeExecutor sample to automatically provision an Agent Engine if neither agent_engine_resource_name nor sandbox_resource_name is provided (6c34694)
• Extract and merge EventActions from A2A metadata (4b677e7), closes #3968
• mcp: add sampling callback support for MCP sessions (8f82697)
• Optional GCP project and credential for GCS access (2f90c1a)
• Support new embedding model in files retrieval (faafac9)

Bug Fixes

• add agent name validation to prevent arbitrary module imports (116f75d)
• add protection for arbitrary module imports (995cd1c), closes #4947
• Add read-only session support in DatabaseSessionService (f6ea58b), closes #4771
• Allow snake case for skill name (b157276)
• bigquery: use valid dataplex OAuth scope (4010716)
• Default to ClusterIP so GKE deployment isn't publicly exposed by default (f7359e3)
• deps: bump google-genai minimum to >=1.64.0 for gemini-embedding-2-preview (f8270c8)
• enforce allowed file extensions for GET requests in the builder API (96e845e)
• error when event does not contain long_running_tool_ids (1f9f0fe)
• Exclude compromised LiteLLM versions from dependencies pin to 1.82.6 (77f1c41)
• Fix IDE hangs by moving test venv and cache to /tmp (6f6fd95)
• Fix imports for environment simulation files (dcccfca)
• gate builder endpoints behind web flag (6c24ccc)
• Handle concurrent creation of app/user state rows in DatabaseSessionService (d78422a), closes #4954
• live: convert response_modalities to Modality enum before assigning to LiveConnectConfig (47aaf66), closes #4869
• models: handle arbitrary dict responses in part_to_message_block (c26d359)
• models: update 429 docs link for Gemini (a231c72)
• populate required for Pydantic BaseModel parameters in FunctionTool (c5d809e), closes #4777
• Prevent compaction of events with pending function calls (991c411), closes #4740
• Prevent uv.lock modifications in unittests.sh (e6476c9)
• Refactor blocking subprocess call to use asyncio in bash_tool (58c4536)
• Refactor LiteLlm check to avoid ImportError (7b94a76)
• Reject appends to stale sessions in DatabaseSessionService (b8e7647), closes #4751
• Remove redundant client_id from fetch_token call (50d6f35), closes #4782
• returns '<No stdout/stderr captured>' instead of empty strings for clearer agent feedback and correct typing (3e00e95)
• Store and retrieve usage_metadata in Vertex AI custom_metadata (b318eee)
• Support resolving string annotations for find_context_parameter (22fc332)
• telemetry: Rolling back change to fix issue affecting LlmAgent creation due to missing version field (0e18f81)
• tools: disable default httpx 5s timeout in OpenAPI tool _request (4c9c01f), closes #4431
• tools: support regional Discovery Engine endpoints (30b904e)
• tools: support structured datastores in DiscoveryEngineSearchTool (f35c3a6), closes #3406
• Update Agent Registry to use the full agent card if available (031f581)
• Update eval extras to Vertex SDK package version with constrained LiteLLM upperbound (27cc98d)
• Update import and version for k8s-agent-sandbox (1ee0623), closes #4883
• Update list_agents to only list directories, not validate agent definitions (5020954)

Code Refactoring

• rename agent simulator to environment simulation. Also add tracing into environment simulation (99a31bf)

Documentation

• Feat/Issue Monitoring Agent (780093f)
• Use a dedicated API key for docs agents (51c19cb)

1.27.5 (2026-03-26)

Bug Fixes

• Update eval extras to Vertex SDK package version with constrained LiteLLM upperbound (77928a8)

1.27.4 (2026-03-24)

Bug Fixes

• Exclude compromised LiteLLM versions from dependencies pin to 1.82.6 (fa5e707)
• gate builder endpoints behind web flag (44b3f72)

1.27.3 (2026-03-23)

Bug Fixes

• add protection for arbitrary module imports (276adfb)

Bug Fixes

• Use valid dataplex OAuth scope for BigQueryToolset (4010716)
• Store and retrieve usage_metadata in Vertex AI custom_metadata (b318eee)

Bug Fixes

• Rolling back change to fix issue affecting LlmAgent creation due to missing version field (0e18f81)

Features

• [Core]

  • Introduce A2A request interceptors in RemoteA2aAgent (6f772d2)
  • Add UiWidget to EventActions for supporting new experimental UI Widgets feature (530ff06)
  • auth: Add pluggable support for auth integrations using AuthProviderRegistry within CredentialManager (d004074)
  • Support all types.SchemaUnion as output_schema in LLM Agent (63f450e)
  • durable runtime support (07fdd23)
  • runners: pass GetSessionConfig through Runner to session service (eff724a)

• [Models]

  • Add support for PDF documents in Anthropic LLM (4c8ba74)
  • Add streaming support for Anthropic models (5770cd3), closes #3250
  • Enable output schema with tools for LiteLlm models (89df5fc), closes #3969
  • Preserve thought_signature in LiteLLM tool calls (ae565be), closes #4650

• [Web]

  • Updated human in the loop: developers now can respond to long running functions directly in chat
  • Render artifacts when resuming
  • Fix some light mode styles
  • Fix token level streaming not working properly (22799c0)

• [Observability]

  • telemetry: add new gen_ai.agent.version span attribute (ffe97ec)
  • otel: add gen_ai.tool.definitions to experimental semconv (4dd4d5e)
  • otel: add experimental semantic convention and emit gen_ai.client.inference.operation.details event (19718e9)
  • add missing token usage span attributes during model usage (77bf325)
  • capture tool execution error code in OpenTelemetry spans (e0a6c6d)

• [Tools]

  • Warn when accessing DEFAULT_SKILL_SYSTEM_INSTRUCTION (35366f4)
  • add preserve_property_names option to OpenAPIToolset (078b516)
  • Add gcs filesystem support for Skills. It supports skills in text and pdf format, also has some sample agents (6edcb97)
  • Add list_skills_in_dir to skills utils (327b3af)
  • Add support for MCP App UI widgets in MCPTool (86db35c)
  • add Dataplex Catalog search tool to BigQuery ADK (82c2eef)
  • Add RunSkillScriptTool to SkillToolset (636f68f)
  • Add support for ADK tools in SkillToolset (44a5e6b)
  • limit number of user-provided BigQuery job labels and reserve internal prefixes (8c4ff74)
  • Add param support to Bigtable execute_sql (5702a4b)
  • bigtable: add Bigtable cluster metadata tools (34c560e)
  • execute-type param addition in GkeCodeExecutor (9c45166)
  • skill: Add BashTool (8a31612)
  • Add support for toolsets to additional_tools field of SkillToolset (066fcec)

• [Optimization]

  • Add adk optimize command (b18d7a1)
  • Add interface between optimization infra and LocalEvalService (7b7ddda)
  • Add GEPA root agent prompt optimizer (4e3e2cb)

• [Integrations]

  • Enhance BigQuery plugin schema upgrades and error reporting (bcf38fa)
  • Enhance BQ plugin with fork safety, auto views, and trace continuity (80c5a24)
  • Handle Conflict Errors in BigQuery Agent Analytics Plugin (372c76b)
  • Added tracking headers for ADK CLI command to Agent Engine (3117446)

• [A2A]

  • New implementation of A2aAgentExecutor and A2A-ADK conversion (87ffc55)
  • New implementation of RemoteA2aAgent and A2A-ADK conversion (6770e41)

Bug Fixes

• Allow artifact services to accept dictionary representations of types.Part (b004da5), closes #2886
• Decode image data from ComputerUse tool response into image blobs (d7cfd8f)
• Expand LiteLLM reasoning extraction to include 'reasoning' field (9468487), closes #3694
• Filter non-agent directories from list_agents() (3b5937f)
• Fix Type Error by initializing user_content as a Content object (2addf6b)
• Handle length finish reason in LiteLLM responses (4c6096b), closes #4482
• In SaveFilesAsArtifactsPlugin, write the artifact delta to state then event actions so that the plugin works with ADK Web UI's artifacts panel (d6f31be)
• Make invocation_context optional in convert_event_to_a2a_message (8e79a12)
• Optimize row-level locking in append_event (d61846f), closes #4655
• Preserve thought_signature in FunctionCall conversions between GenAI and A2A (f9c104f)
• Prevent splitting of SSE events with artifactDelta for function resume requests (6a929af), closes #4487
• Propagate file names during A2A to/from Genai Part conversion (f324fa2)
• Propagate thought from A2A TextPart metadata to GenAI Part (e59929e)
• Re-export DEFAULT_SKILL_SYSTEM_INSTRUCTION to skills and skill/prompt.py to avoid breaking current users (de4dee8)
• Refactor type string update in Anthropic tool param conversion (ab4b736)
• simulation: handle NoneType generated_content (9d15517)
• Store and retrieve EventCompaction via custom_metadata in Vertex AISessionService (2e434ca), closes #3465
• Support before_tool_callback and after_tool_callback in Live mode (c36a708), closes #4704
• temp-scoped state now visible to subsequent agents in same invocation (2780ae2)
• tools: Handle JSON Schema boolean schemas in Gemini schema conversion (3256a67)
• typo in A2A EXPERIMENTAL warning (eb55eb7)
• Update agent_engine_sandbox_code_executor in ADK (dff4c44)
• update Bigtable query tools to async functions (72f3e7e)
• Update expected UsageMetadataChunk in LiteLLM tests (dd0851a), closes #4680
• update toolbox server and SDK package versions (2e370ea)
• Validate session before streaming instead of eagerly advancing the runner generator (ebbc114)

Code Refactoring

• extract reusable functions from hitl and auth preprocessor (c59afc2)
• Rename base classes and TypeVars in optimization data types (9154ef5)

1.26.0 (2026-02-26)

Features

• [Core]

  • Add intra-invocation compaction and token compaction pre-request (485fcb8)
  • Use --memory_service_uri in ADK CLI run command (a7b5097)

• [Models]

  • Add /chat/completions integration to ApigeeLlm (9c4c445)
  • Add /chat/completions streaming support to Apigee LLM (121d277)
  • Expand LiteLlm supported models and add registry tests (d5332f4)

• [Tools]

  • Add load_skill_from_dir() method (9f7d5b3)
  • Agent Skills spec compliance — validation, aliases, scripts, and auto-injection (223d9a7)
  • BigQuery ADK support for search catalog tool (bef3f11)
  • Make skill instruction optimizable and can adapt to user tasks (21be6ad)
  • Pass trace context in MCP tool call's _meta field with OpenTelemetry propagator (bcbfeba)

• [Evals]

  • Introduce User Personas to the ADK evaluation framework (6a808c6)

• [Services]

  • Add generate/create modes for Vertex AI Memory Bank writes (811e50a)
  • Add support for memory consolidation via Vertex AI Memory Bank (4a88804)

• [A2A]

  • Add interceptor framework to A2aAgentExecutor (87fcd77)

• [Auth]

  • Add native support for id_token in OAuth2 credentials (33f7d11)
  • Support ID token exchange in ServiceAccountCredentialExchanger (7be90db), closes #4458

• [Integrations]

  • Agent Registry in ADK (abaa929)
  • Add schema auto-upgrade, tool provenance, HITL tracing, and span hierarchy fix to BigQuery Agent Analytics plugin (4260ef0)
  • Change default BigQuery table ID and update docstring (7557a92)
  • Update Agent Registry to create AgentCard from info in get agents endpoint (c33d614)

• [Web]

  • Enable dependency injection for agent loader in FastAPI app gen (34da2d5)

Bug Fixes

• Add OpenAI strict JSON schema enforcement in LiteLLM (2dbd1f2), closes #4573
• Add push notification config store to agent_to_a2a (4ca904f), closes #4126
• Add support for injecting a custom google.genai.Client into Gemini models (48105b4), closes #2560
• Add support for injecting a custom google.genai.Client into Gemini models (c615757), closes #2560
• Check both input_stream parameter name and its annotation to decide whether it's a streaming tool that accept input stream (d56cb41)
• deps: Increase pydantic lower version to 2.7.0 (dbd6420)
• edit copybara and BUILD config for new adk/integrations folder (added with Agent Registry) (37d52b4)
• Expand add_memory to accept MemoryEntry (f27a9cf)
• Fix pickling lock errors in McpSessionManager (4e2d615)
• fix typo in PlanReActPlanner instruction (6d53d80)
• handle UnicodeDecodeError when loading skills in ADK (3fbc27f)
• Improve BigQuery Agent Analytics plugin reliability and code quality (ea03487)
• Include list of skills in every message and remove list_skills tool from system instruction (4285f85)
• Invoke on_tool_error_callback for missing tools in live mode (e6b601a)
• Keep query params embedded in OpenAPI paths when using httpx (ffbcc0a), closes #4555
• Only relay the LiveRequest after tools is invoked (b53bc55)
• Parallelize tool resolution in LlmAgent.canonical_tools() (7478bda)
• race condition in table creation for DatabaseSessionService (fbe9ecc)
• Re-export DEFAULT_SKILL_SYSTEM_INSTRUCTION to skills and skill/prompt.py to avoid breaking current users (40ec134)
• Refactor LiteLLM streaming response parsing for compatibility with LiteLLM 1.81+ (e8019b1), closes #4225
• remove duplicate session GET when using API server, unbreak auto_session_create when using API server (445dc18)
• Remove experimental decorators from user persona data models (eccdf6d)
• Replace the global DEFAULT_USER_PERSONA_REGISTRY with a function call to get_default_persona_registry (2703613)
• skill: coloate default skill SI with skilltoolset (fc1f1db)
• Update agent_engine_sandbox_code_executor in ADK (ee8d956)
• Update agent_engine_sandbox_code_executor in ADK (dab80e4)
• Update sample skills agent to use weather-skill instead of weather_skill (8f54281)
• update Spanner query tools to async functions (1dbcecc)
• use correct msg_out/msg_err keys for Agent Engine sandbox output (b1e33a9)
• Validate session before streaming instead of eagerly advancing the runner generator (ab32f33)
• web: allow session resume without new message (30b2ed3)

Code Refactoring

• Extract reusable function for building agent transfer instructions (e1e0d63)
• Extract reusable private methods (976a238)
• Extract reusable private methods (42eeaef)
• Extract reusable private methods (706f9fe)

Documentation

• add thinking_config in generate_content_config in example agent (c6b1c74)

1.25.1 (2026-02-18)

Bug Fixes

• Fix pickling lock errors in McpSessionManager (4e2d615)

1.25.0 (2026-02-11)

Features

• [Core]

  • Add a demo for the simple prompt optimizer for the optimization interface (0abf4cd)
  • Add --auto_create_session flag to adk api_server CLI (40c15d0)
  • Add add_events_to_memory facade for event-delta (59e8897)
  • Add post-invocation token-threshold compaction with event retention (a88e864)
  • Add report generation to adk conformance test command (43c437e)

• [Models]

  • Add base_url option to Gemini LLM class (781f605)

• [Tools]

  • Enhance google credentials config to support externally passed access token (3cf43e3)
  • Update agent simulator by improving prompts and add environment data (7af1858)
  • Add a load MCP resource tool (e25227d)
  • Add SkillToolset to adk (8d02792)

• [Web]

  • Add /health and /version endpoints to ADK web server (25ec2c6)

Bug Fixes

• Use async iteration for VertexAiSessionService.list_sessions pagination (758d337)
• Fix event loop closed bug in McpSessionManager (4aa4751)
• Preserve thought_signature in function call conversions for interactions API integration (2010569)
• Propagate grounding and citation metadata in streaming responses (e6da417)
• Add endpoints to get/list artifact version metadata (e0b9712)
• Support escaped curly braces in instruction templates (7c7d25a)
• Strip timezone for PostgreSQL timestamps in DatabaseSessionService (19b6076)
• Prompt token may be None in streaming mode (32ee07d)
• Pass invocation_id from /run endpoint to Runner.run_async (d2dba27)
• Conditionally preserve function call IDs in LLM requests (663cb75)
• Migrate VertexAiMemoryBankService to use the async Vertex AI client (64a44c2)
• Handle list values in Gemini schema sanitization (fd8a9e3)
• Used logger to log instead of print in MCP (6bc70a6)

Improvements

• Replace check of instance for LlmAgent with hasAttribute check (7110336)
• Log exception details before re-raising in MCP session execution (de79bf1)

1.24.1 (2026-02-06)

Bug Fixes

• Add back deprecated eval endpoint for web until we migrate(ae993e8)
• Update eval dialog colors, and fix a2ui component types (3686a3a)

1.24.0 (2026-02-04)

⚠ BREAKING CHANGES

• Breaking: Make credential manager accept tool_context instead of callback_context (fe82f3c)

Highlights

• [Web]
  • Consolidated Event View: Replaced the Event tab with a more intuitive "click-to-expand" interaction on message rows, enabling faster debugging within the chat context
  • Enhanced Accessibility: Added full support for arrow-key navigation for a more seamless, keyboard-centric experience
  • Rich Developer Tooling: Introduced detailed tooltips for function calls, providing instant visibility into arguments, responses, and state changes
  • A2UI Integration: Integrated the A2UI v0.8 standard catalog to automatically render spec-compliant ADK parts as native UI components directly in the chat

Features

• [Core]

  • Allow passthrough of GOOGLE_CLOUD_LOCATION for Agent Engine deployments (004e15c)
  • Add interface for agent optimizers (4ee125a)
  • Pass event ID as metadata when converted into a message (85434e2)
  • Restructure the bug report template as per the intake process (324796b)

• [Models]

  • Mark Vertex calls made from non-Gemini models (7d58e0d)

• [Evals]

  • Allow Vertex AI Client initialization with API Key (43d6075)
  • Remove overall evaluation status calculation from _CustomMetricEvaluator and add threshold to custom metric function expected signature (553e376)

• [Tools]

  • Make OpenAPI tool asynchronous (9290b96)
  • Implement toolset authentication for McpToolset, OpenAPIToolset, and other toolsets (798f65d)
  • Add framework support for toolset authentication before get_tools calls (ee873ca)
  • Support dynamic configuration for VertexAiSearchTool (585ebfd)
  • Add get_auth_config method to toolset to expose authentication requirements (381d44c)
  • Add methods in McpToolset for users to access MCP resources (8f7d965)
  • Improve error message when failing to get tools from MCP (3480b3b)

• [Services]

  • Improve asyncio loop handling and test cleanup (00aba2d)

• [Live]

  • Support running tools in separate threads for live mode (714c3ad)

• [Observability]

  • Add extra attributes to spans generated with opentelemetry-instrumentation-google-genai (e87a843)

Bug Fixes

• Ignore session_db_kwargs for SQLite session services (ce07cd8)
• Resolve MutualTLSChannelError by adding pyopenssl dependency (125bc85)
• Add update_timestamp_tz property to StorageSession (666cebe)
• Do not treat Function Calls and Function Responses as invisible when marked as thoughts (853a3b0)
• Add pre-deployment validation for agent module imports (credit to @ppgranger, 2ac468e)
• Fix cases where execution_result_delimiters have None type element (a16e3cc)
• Disable save_input_blobs_as_artifacts deprecation warning message for users not setting it (c34615e)
• Fix agent config path handling in generated deployment script (8012339)
• Add pypika>=0.50.0 to project.toml to support crewai on Python 3.12+ (e8f7aa3)
• Update OpenTelemetry dependency versions to relax version constraints for opentelemetry-api and opentelemetry-sdk (706a6dd)
• Enable pool_pre_ping by default for non-SQLite database engines (da73e71)
• Ensure database sessions are always rolled back on errors (63a8eba)
• Reload stale session in DatabaseSessionService when storage update time is later than the in-memory session object (1063fa5)
• Make credential key generation stable and prevent cross-user credential leaks (33012e6)
• Change MCP read_resource to return original contents (ecce7e5)
• Recognize function responses as non-empty parts in LiteLLM (d0102ec)
• Handle HTTP/HTTPS URLs for media files in LiteLLM content conversion (47221cd)
• Fix Pydantic schema generation error for ClientSession (131fbd3)
• Fix Click’s Wrapping in adk eval help message (3bcd8f7)
• Stream errors as simple JSON objects in ADK web server SSE endpoint (798d005)
• Remove print debugging artifact (0d38a36)

Improvements

• Check will_continue for streaming function calls (2220d88)
• Update ADK web, rework events, and add A2UI capabilities (37e6507)
• Improve error handling for LiteLLM import in gemma_llm.py (574ec43)
• Replace proxy methods with utils implementation (6ff10b2, f82ceb0)
• Replace print statements with logging in ADK evaluation components (dd8cd27)
• Add sample agent that requires OAuth flow during MCP tool listing, and convert MCPToolset to McpToolset in unit tests (2770012, 4341839)
• Ensure BigQueryAgentAnalyticsPlugin is shut down after each test (c0c98d9)
• Add ADK logger in RestApiTool (288c2c4)
• Add GitHub Action check to run mypy (32f9f92)
• Add unittests.sh script and update CONTRIBUTING.md (025b42c)
• Extract helper function for LLM request building and response processing (753084f)

1.23.0 (2026-01-22)

⚠ BREAKING CHANGES

• Breaking: Use OpenTelemetry for BigQuery plugin tracing, replacing custom ContextVar implementation (ab89d12)

Features

• [Core]

  • Add support to automatically create a session if one does not exist (8e69a58)
  • Remove @experimental decorator from AgentEngineSandboxCodeExecutor (135f763)
  • Add --disable_features CLI option to override default feature enable state (53b67ce)
  • Add otel_to_cloud flag to adk deploy agent_engine command (21f63f6)
  • Add is_computer_use field to agent information in adk-web server (5923da7)
  • Allow thinking_config in generate_content_config (e162bb8)
  • Convert A2UI messages between A2A DataPart metadata and ADK events (1133ce2)
  • Add --enable_features CLI option to override default feature enable state (79fcddb)

• [Tools]

  • Add flush mechanism to BigQueryAgentAnalyticsPlugin to ensure pending log events are written to BigQuery (9579bea)
  • Allow Google Search tool to set a different model (b57a3d4)
  • Support authentication for MCP tool listing (e3d542a 19315fe)
  • Use JSON schema for base_retrieval_tool, load_artifacts_tool, and load_memory_tool declarations when the feature is enabled (69ad605)
  • Use JSON schema for IntegrationConnectorTool declaration when the feature is enabled (2ed6865)
  • Start and close ClientSession in a single task in McpSessionManager (cce430d)
  • Use JSON schema for RestApiTool declaration when the feature is enabled (a5f0d33)

• [Evals]

  • Update adk eval CLI to consume custom metrics by adding CustomMetricEvaluator (ea0934b)
  • Update EvalConfig and EvalMetric data models to support custom metrics (6d2f33a)

• [Observability]

  • Add minimal generate_content {model.name} spans and logs for non-Gemini inference and when opentelemetry-inference-google-genai dependency is missing (935c279)

• [Integrations]

  • Enhance TraceManager asynchronous safety, enrich BigQuery plugin logging, and fix serialization (a4116a6)

• [Live]

  • Persist user input content to session in live mode (a04828d)

Bug Fixes

• Recursively extract input/output schema for AgentTool (bf2b56d)
• Yield buffered function_call and function_response events during live streaming (7b25b8f)
• Update authlib and mcp dependency versions (7955177)
• Set LITELLM_MODE to PRODUCTION before importing LiteLLM to prevent implicit .env file loading (215c2f5)
• Redact sensitive information from URIs in logs (5257869)
• Handle asynchronous driver URLs in the migration tool (4b29d15)
• Remove custom metadata from A2A response events (81eaeb5)
• Handle None inferences in eval results (7d4326c)
• Mark all parts of a thought event as thought (f92d4e3)
• Use json.dumps for error messages in SSE events (6ad18cc)
• Use the correct path for config-based agents when deploying to AgentEngine (83d7bb6)
• Support Generator and Async Generator tool declarations in JSON schema (19555e7 7c28297)
• Prevent stopping event processing on events with None content (ed2c3eb)
• Fix 'NoneType' object is not iterable error (7db3ce9)
• Use canonical tools to find streaming tools and register them by tool.name (ec6abf4)
• Initialize self._auth_config inside BaseAuthenticatedTool to access authentication headers in McpTool (d4da1bb)
• Only filter out audio content when sending history (712b5a3)
• Add finish reason mapping and remove custom file URI handling in LiteLLM (89bed43)
• Convert unsupported inline artifact MIME types to text in LoadArtifactsTool (fdc98d5)
• Pass log_level to uvicorn in web and api_server commands (38d52b2)
• Use the agent name as the author of the audio event (ab62b1b)
• Handle NOT_FOUND error when fetching Vertex AI sessions (75231a3)
• Fix httpx client closure during event pagination (b725045)

Improvements

• Add new conversational analytics API toolset (82fa10b)
• Filter out adk_request_input event from content list (295b345)
• Always skip executing partial function calls (d62f9c8)
• Update comments of request confirmation preprocessor (1699b09)
• Fix various typos (a8f2ddd)
• Update sample live streaming tools agent to use latest live models (3dd7e3f)
• Make the regex to catch CLI reference strict by adding word boundary anchor (c222a45)
• Migrate ToolboxToolset to use toolbox-adk and align validation (7dc6adf 277084e)
• Always log API backend when connecting to live model (7b035aa)
• Add a sample BigQuery agent using BigQuery MCP tools (672b57f)
• Add a DebugLoggingPlugin to record human-readable debugging logs (8973618)
• Upgrade the sample BigQuery agent model version to gemini-2.5-flash (fd2c0f5)
• Import migration_runner lazily within the migrate command (905604f)

Bug Fixes

• Add back adk migrate session CLI (8fb2be2).
• Escape database reserved keyword (94d48fc).

Features

• [Core]

  • Make LlmAgent.model optional with a default fallback (b287215).
  • Support regex for allowed origins (2ea6e51).
  • Enable PROGRESSIVE_SSE_STREAMING feature by default (0b1cff2).

• [Evals]

  • Add custom instructions support to LlmBackedUserSimulator (a364388).
  • Introduce a post-hoc, per-turn evaluator for user simulations (e515e0f).

• [Tools]

  • Expose mcps streamable http custom httpx factory parameter (bfed19c).
  • Add a handwritten tool for Cloud Pub/Sub (b6f6dcb).
  • Add token_endpoint_auth_method support to OAuth2 credentials (8782a69).

• [Services]

  • Introduce new JSON-based database schema for DatabaseSessionService, which will be used for newly-created databases. A migration command and script are provided.(7e6ef71 ba91fea ce64787).
  • Set log level when deploying to Agent Engine (1f546df).

• [A2A]

  • Update event_converter used in A2ARemote agent to use a2a_task.status.message only if parts are non-empty (e4ee9d7).

Bug Fixes

• Add checks for event content and parts before accessing (5912835).
• Validate app name in adk create command (742c926).
• Prevent .env files from overriding existing environment variables (0827d12).
• Prevent ContextFilterPlugin from creating orphaned function responses (e32f017).
• Update empty event check to include executable code and execution results (688f48f).
• Make the BigQuery analytics plugin work with agents that don't have instructions such as the LoopAgent (8bed01c).
• Label response as thought if task is immediately returned as working (4f3b733).
• Move and enhance the deprecation warning for the plugins argument in "_validate_runner_params" to the beginning of the function (43270bc).
• Oauth refresh not triggered on token expiry (69997cd).
• Fix double JSON encoding when saving eval set results (fc4e3d6).
• Allow string values for ToolTrajectoryCriterion.match_type (93d6e4c).
• Fix inconsistent method signatures for evaluate_invocations (0918b64).
• Honor the modalities parameter in adk api server for live API (19de45b).
• Filter out thought parts in lite_llm._get_content (1ace8fc).
• Rehydration of EventActions in StorageEvent.to_event (838530e).
• Heal missing tool results before LiteLLM requests (6b7386b).
• Refine Ollama content flattening and provider checks (c6f389d).
• Add MIME type inference and default for file URIs in LiteLLM (5c4bae7).
• Use mode='json' in model_dump to serialize bytes correctly when using telemetry (96c5db5).
• Avoid local .adk storage in Cloud Run/GKE (b30c2f4).
• Remove fallback to cached exchanged credential in _load_existing_credential (1ae0e16).
• Handle overriding of requirements when deploying to agent engine (38a30a4).
• Built-in agents (names starting with "__") now use in-memory session storage instead of creating .adk folders in the agents directory (e3bac1a).
• Change error_message column type to TEXT in DatabaseSessionService (8335f35).
• Add schema type sanitization to OpenAPI spec parser (6dce7f8).
• Prevent retry_on_errors from retrying asyncio.CancelledError (30d3411).
• Include back-ticks around the BQ asset names in the tools examples (8789ad8).
• Fix issue with MCP tools throwing an error (26e77e1).
• Exclude thought parts when merging agent output (07bb164).
• Prepend "https://" to the MCP server url only if it doesn't already have a scheme (71b3289).
• Split SSE events with both content and artifactDelta in ADK Web Server (084fcfa).
• Propagate RunConfig custom metadata to all events (e3db2d0).
• Harden YAML builder tmp save/cleanup(6f259f0).
• Ignore adk-bot administrative actions in stale agent (3ec7ae3).
• Only prepend "https://" to the MCP server url if it doesn't already have a scheme (71b3289).
• Check all content parts for emptiness in _contains_empty_content (f35d129).

Improvements

• Remove unnecessary event loop creation in LiveRequstQueue constructor (ecc9f18).
• Close database engines to avoid aiosqlite pytest hangs (4ddb2cb).
• Add override_feature_enabled to override the default feature enable states (a088506).
• Move SQLite migration script to migration/ folder (e8ab7da).
• Update latest Live Model names for sample agent (f1eb1c0).
• Update google-genai and google-cloud-aiplatform versions (d58ea58).
• Introduce MetricInfoProvider interface, and refactor metric evaluators to use this interface to provide MetricInfo (5b7c8c0).
• Update _flatten_ollama_content return type and add tests (fcea86f).
• Add disambiguation message to enterprise_search_tool (8329fec).
• Add x-goog-user-project header to http calls in API Registry (0088b0f).
• Set the default response modality to AUDIO only (a4b914b).