This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+10 more
ReleasePort's take
Moderate signalVersion v2.1.0 adds chart generation and artifact loading to the data agent, supports sandbox creation from templates/snapshots, and includes user.id in gen_ai telemetry logs.
Why it matters: These feature additions enable richer analytics (charting), streamlined environment provisioning (sandbox templates/snapshots), and improved observability of AI interactions via user‑id logging.
Summary
AI summaryAdd chart generation, sandbox creation from templates/snapshots, user.id telemetry field in gen_ai logs.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Adds chart generation and artifact loading to data agent Adds chart generation and artifact loading to data agent Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Feature | Medium |
Adds support for creating sandboxes from templates and snapshots Adds support for creating sandboxes from templates and snapshots Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Feature | Medium |
Adds user.id to gen_ai.user.message log records for telemetry Adds user.id to gen_ai.user.message log records for telemetry Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Dependency | Medium |
Adds python-dateutil dependency to stale-bot CI workflow Adds python-dateutil dependency to stale-bot CI workflow Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Performance | Medium |
Lazy‑imports GCS evaluation managers in evals utility Lazy‑imports GCS evaluation managers in evals utility Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Performance | Medium |
Makes google‑cloud‑storage import lazy in skill utils Makes google‑cloud‑storage import lazy in skill utils Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Bugfix | Medium |
Fixes error message telemetry for tool calls Fixes error message telemetry for tool calls Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Preserves transcription event order in conversation trajectory Preserves transcription event order in conversation trajectory Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Fixes --reload_agents behavior for web CLI Fixes --reload_agents behavior for web CLI Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Converts Union[Pydantic, Pydantic] tool arguments at runtime Converts Union[Pydantic, Pydantic] tool arguments at runtime Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Fixes input and output transcription finished events for Gemini v3.1 Fixes input and output transcription finished events for Gemini v3.1 Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Resolves circular import caused by llm_request Resolves circular import caused by llm_request Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Resolves circular import in base_tool Resolves circular import in base_tool Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Prevents session drop on MCP tool error Prevents session drop on MCP tool error Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Updates EditFileTool to handle cross‑platform line breaks and escape regex characters Updates EditFileTool to handle cross‑platform line breaks and escape regex characters Source: llm_adapter@2026-05-23 Confidence: high |
— |
| Bugfix | Medium |
Prevents workflow failures in relocated adk_team samples CI Prevents workflow failures in relocated adk_team samples CI Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Bugfix | Medium |
Uses absolute path for PYTHONPATH in upload docs CI workflow Uses absolute path for PYTHONPATH in upload docs CI workflow Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Bugfix | Medium |
Informs user to install optional google.cloud dependency when missing Informs user to install optional google.cloud dependency when missing Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Bugfix | Medium |
Fixes bug where grounding metadata in Gemini 3.1 live was silently discarded Fixes bug where grounding metadata in Gemini 3.1 live was silently discarded Source: llm_adapter@2026-05-23 Confidence: low |
— |
| Bugfix | Medium |
Appends trailing newline to JSON test outputs in tests suite Appends trailing newline to JSON test outputs in tests suite Source: llm_adapter@2026-05-23 Confidence: low |
— |
Full changelog
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)
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About adk-python
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
Related context
Related tools
Beta — feedback welcome: [email protected]