release v1.6.1 at 2026-09-16 12:22:16 UTC

Author: pvlsirotkin

.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "google-agents-cli",
-  "version": "1.5.0",
+  "version": "1.6.1",
   "description": "Scaffold, develop, evaluate, and deploy AI agents with Google ADK. Bundles skills for the agent development lifecycle.",
   "author": { "name": "Google LLC" },
   "homepage": "https://github.com/google/agents-cli",

RELEASE_NOTES.md

@@ -3,6 +3,38 @@
 All notable changes to this project will be documented in this file.
 
 
+## [1.6.1] - 2026-09-16
+
+- **ADK Go is now available to everyone (Preview).** The `adk_go` template ("Go agent with A2A") now appears in `agents-cli scaffold create` with no experiment override, and Go projects get the toolchain Python already has:
+  - Deploy to Agent Runtime, Cloud Run and GKE, with session methods exposed to `agent_engines.get()` clients.
+  - CloudSQL, Agent Platform and in-memory session services.
+  - Prompts and responses in BigQuery Agent Analytics, and telemetry through OTLP.
+  - CI/CD templates at parity with Python.
+  - The bundled skills cover Go and Python.
+- Scaffolds now default to `gemini-3.8-flash`.
+- Pin `google-adk` below 2.9.0. On 2.9.0 a Cloud Run agent's prompts and responses stopped reaching BigQuery Agent Analytics, leaving `completions_view` empty.
+- `agents-cli scaffold create` no longer asks for a region when the deployment target is `none`.
+- The one-off `agents-cli run` no longer prints stop and resume hints for a server it did not leave running.
+- `agents-cli deploy` prints how to talk to a Cloud Run agent once the deploy finishes.
+- `agents-cli deploy` can change the identity type on an existing agent.
+- `agents-cli deploy` now reports the correct identity after a deploy, instead of the wrong principal for agents using Agent Identity.
+- Exclude vendored extensions from the deploy payload, where they could add megabytes to every deploy.
+- `agents-cli infra single-project` now correctly enables the Service Usage API, so a new project bootstraps without enabling it by hand.
+- Fix remote ADK samples generating two agent directories and an unbuildable project.
+- Scaffolded Dockerfiles install uv from a pinned image instead of `pip install`.
+- `agents-cli run` and `agents-cli eval` no longer stop a background server they only borrowed.
+  - https://github.com/google/agents-cli/issues/71
+- Startup and version notices now go to stderr, so `--json` output stays machine-readable.
+  - https://github.com/google/agents-cli/issues/69
+- `agents-cli eval generate` no longer silently drops per-case session state.
+  - https://github.com/google/agents-cli/issues/52
+- `agents-cli run --mode a2a` now shows status updates, errors and empty turns instead of staying silent.
+- The `run` resume hint includes `--url` and the other routing flags.
+- `agents-cli info --json` reports `root_agent_name`.
+- `agents-cli extension add` no longer copies ignored files from a `local@` source.
+- Fix the macOS-only `sed` in the Starter Pack migration guide and spell out the dry-run and apply steps.
+  - https://github.com/google/agents-cli/issues/78
+
 ## [1.5.0] - 2026-09-01
 
 - **Extension system with a LangChain template.** `agents-cli` can now be extended with custom commands and add-on agent templates, shipped with a LangChain extension and authoring guides. Extensions install from GitHub, other git hosts (including self-hosted and enterprise servers), or a local path for development.

docs/pyproject.toml

@@ -19,7 +19,7 @@ description = "Documentation site for agents-cli"
 requires-python = ">=3.11"
 dependencies = [
     "mkdocs==1.6.1",
-    "mkdocs-material==9.7.6",
+    "mkdocs-material==9.7.7",
     "mkdocs-click==0.9.0",
     "google-agents-cli",
 ]

docs/src/guide/evaluation.md

@@ -94,13 +94,13 @@ custom_metrics:
       Final response: {response}
       Full trace (for tool-call and reasoning context): {agent_data}
       Return JSON: {"score": <1|2|3|4|5>, "explanation": "<reason>"}
-    judge_model: gemini-3.7-flash
+    judge_model: gemini-3.8-flash
     judge_model_sampling_count: 3
 ```
 
 Each custom metric must conform to either the **Code Execution Metric** or **LLM-as-a-Judge Metric** (`LLMMetric`) schema:
 - **Code Execution Metric**: Used to run custom Python code for evaluation. Must have a `name` and a `custom_function` (containing a `def evaluate(instance):` signature). By default, the function executes **locally in the CLI process** — no GCP project or region is required, but the user-supplied code runs with the CLI's privileges. Add `"execution": "remote"` to opt into Vertex AI's sandboxed `CodeExecutionMetric` (server-side), which requires a configured GCP project + region.
-- **LLM-as-a-Judge Metric**: Used to evaluate responses using an LLM judge. Must have a `name` and a `prompt_template`. Optional fields include `judge_model` (e.g., `gemini-3.7-flash`) and `judge_model_sampling_count` (between `1` and `32`).
+- **LLM-as-a-Judge Metric**: Used to evaluate responses using an LLM judge. Must have a `name` and a `prompt_template`. Optional fields include `judge_model` (e.g., `gemini-3.8-flash`) and `judge_model_sampling_count` (between `1` and `32`).
 
 ### Quick Reference for Common Scenarios
 

docs/src/guide/extensions/first-party.md

@@ -24,7 +24,7 @@ The LangChain template runs a [LangChain](https://python.langchain.com) agent th
 
 ```bash
 agents-cli create my-agent \
-    --agent google/agents-cli/extensions/langchain/template@v1.5.0 \
+    --agent google/agents-cli/extensions/langchain/template@v1.6.1 \
     -d cloud_run
 cd my-agent && agents-cli install
 ```
@@ -66,7 +66,7 @@ AGENTS_CLI_DISABLE_OVERRIDES=1 agents-cli run --url https://<service-url> --mode
 ### Full journey
 
 ```bash
-agents-cli create my-agent --agent google/agents-cli/extensions/langchain/template@v1.5.0 -d cloud_run
+agents-cli create my-agent --agent google/agents-cli/extensions/langchain/template@v1.6.1 -d cloud_run
 cd my-agent && agents-cli install
 agents-cli run "what's the weather in San Francisco?"       # in-process graph
 agents-cli eval generate --dataset tests/eval/datasets/basic-dataset.json -o tests/eval/output/

docs/src/guide/hands-on-tutorial.md

@@ -62,7 +62,7 @@ The important file is `app/agent.py`. Open it and you'll see two tool functions
 ```python title="app/agent.py"
 root_agent = Agent(
     name="root_agent",
-    model=Gemini(model="gemini-3.7-flash"),
+    model=Gemini(model="gemini-3.8-flash"),
     instruction="You are a helpful AI assistant designed to provide accurate and useful information.",
     tools=[get_weather, get_current_time],
 )
@@ -111,7 +111,7 @@ Let's give the agent a personality. Open `app/agent.py` and change the instructi
 root_agent = Agent(
     name="root_agent",
     model=Gemini(
-        model="gemini-3.7-flash",
+        model="gemini-3.8-flash",
         retry_options=types.HttpRetryOptions(attempts=3),
     ),
     instruction="""You are a cheerful weather reporter who speaks in short, 

docs/src/guide/lifecycle.md

@@ -95,7 +95,7 @@ from google.adk.models import Gemini
 
 root_agent = Agent(
     name="root_agent",
-    model=Gemini(model="gemini-3.7-flash"),
+    model=Gemini(model="gemini-3.8-flash"),
     instruction="You are an SRE outage-recovery assistant...",
     tools=[query_logs, check_metrics, search_runbook],
 )

docs/src/guide/project-structure.md

@@ -50,7 +50,7 @@ from google.adk.apps import App
 from google.adk.models import Gemini
 from google.genai import types
 
-MODEL = "gemini-3.7-flash"
+MODEL = "gemini-3.8-flash"
 
 
 def get_weather(query: str) -> str:
@@ -87,7 +87,7 @@ The four key parts:
 1. **Tool functions** — plain Python functions with docstrings. The docstring tells the LLM when to use the tool.
 2. **`Agent`** — combines a model, instruction (system prompt), and tools.
 3. **`App`** — wraps the agent for serving. The `name` must match the directory name (`app`).
-4. **Model** — defaults to `gemini-3.7-flash`. Change it via the `MODEL` constant at the top of `agent.py`.
+4. **Model** — defaults to `gemini-3.8-flash`. Change it via the `MODEL` constant at the top of `agent.py`.
 
 ### `pyproject.toml`
 

docs/src/guide/quickstart-tutorial.md

@@ -54,7 +54,7 @@ The agent definition ends up looking something like:
 ```python title="app/agent.py"
 root_agent = Agent(
     name="caveman_agent",
-    model=Gemini(model="gemini-3.7-flash"),
+    model=Gemini(model="gemini-3.8-flash"),
     instruction="""You caveman compressor. Human give long words,
     you make short. Rules:
     - No articles. No filler. No fluff.

docs/src/javascripts/lifecycle.js

@@ -840,18 +840,18 @@
   //   execute_tool <tool.name>             (per trace_tool_call)
   const TRACE_SPANS = [
     { id: "01", parent: null, name: "invoke_agent orchestrator", start: 0, dur: 5800, attrs: { "gen_ai.conversation.id": "sess_3f0b…", "gen_ai.agent.name": "orchestrator", "gen_ai.operation.name": "invoke_agent" } },
-    { id: "02", parent: "01", name: "generate_content gemini-3.7-flash", start: 30, dur: 320, attrs: { "gen_ai.usage.input_tokens": 220, "gen_ai.usage.output_tokens": 38 } },
+    { id: "02", parent: "01", name: "generate_content gemini-3.8-flash", start: 30, dur: 320, attrs: { "gen_ai.usage.input_tokens": 220, "gen_ai.usage.output_tokens": 38 } },
     { id: "10", parent: "01", name: "invoke_agent investigator", start: 360, dur: 1900, attrs: { "gen_ai.agent.name": "investigator", "gen_ai.operation.name": "invoke_agent" } },
-    { id: "11", parent: "10", name: "generate_content gemini-3.7-flash", start: 380, dur: 220, attrs: { "gen_ai.usage.input_tokens": 320, "gen_ai.usage.output_tokens": 42 } },
+    { id: "11", parent: "10", name: "generate_content gemini-3.8-flash", start: 380, dur: 220, attrs: { "gen_ai.usage.input_tokens": 320, "gen_ai.usage.output_tokens": 42 } },
     { id: "12", parent: "10", name: "execute_tool query_logs", start: 610, dur: 720, attrs: { "gen_ai.tool.name": "query_logs", "gcp.vertex.agent.tool_call_args": '{"service":"api-gateway"}' } },
     { id: "13", parent: "10", name: "execute_tool check_metrics", start: 1340, dur: 540, attrs: { "gen_ai.tool.name": "check_metrics", "gcp.vertex.agent.tool_call_args": '{"service":"auth-service"}' } },
     { id: "20", parent: "01", name: "invoke_agent diagnoser", start: 2280, dur: 1700, attrs: { "gen_ai.agent.name": "diagnoser", "gen_ai.operation.name": "invoke_agent" } },
-    { id: "21", parent: "20", name: "generate_content gemini-3.7-flash", start: 2300, dur: 280, attrs: { "gen_ai.usage.input_tokens": 412, "gen_ai.usage.output_tokens": 64 } },
+    { id: "21", parent: "20", name: "generate_content gemini-3.8-flash", start: 2300, dur: 280, attrs: { "gen_ai.usage.input_tokens": 412, "gen_ai.usage.output_tokens": 64 } },
     { id: "22", parent: "20", name: "execute_tool search_runbook", start: 2590, dur: 1260, attrs: { "gen_ai.tool.name": "search_runbook", "gcp.vertex.agent.tool_call_args": '{"query":"api-gateway 5xx auth-service latency"}' } },
     { id: "30", parent: "01", name: "invoke_agent remediator", start: 4000, dur: 1280, attrs: { "gen_ai.agent.name": "remediator", "gen_ai.operation.name": "invoke_agent" } },
-    { id: "31", parent: "30", name: "generate_content gemini-3.7-flash", start: 4020, dur: 290, attrs: { "gen_ai.usage.input_tokens": 380, "gen_ai.usage.output_tokens": 58 } },
+    { id: "31", parent: "30", name: "generate_content gemini-3.8-flash", start: 4020, dur: 290, attrs: { "gen_ai.usage.input_tokens": 380, "gen_ai.usage.output_tokens": 58 } },
     { id: "32", parent: "30", name: "execute_tool propose_remediation", start: 4320, dur: 580, attrs: { "gen_ai.tool.name": "propose_remediation", "gcp.vertex.agent.tool_call_args": '{"action":"traffic-shift"}' } },
-    { id: "40", parent: "01", name: "generate_content gemini-3.7-flash", start: 5300, dur: 480, attrs: { "gen_ai.usage.input_tokens": 680, "gen_ai.usage.output_tokens": 92 } },
+    { id: "40", parent: "01", name: "generate_content gemini-3.8-flash", start: 5300, dur: 480, attrs: { "gen_ai.usage.input_tokens": 680, "gen_ai.usage.output_tokens": 92 } },
   ];
 
   function initTrace(host) {