agent_engine_python_deployer

Name: agent_engine_python_deployer
Author: VeerMuchandi

15stars1forks

How to install agent_engine_python_deployer?

veermuchandi/rad-skills/agent_engine_python_deployer

$npx -y skills add VeerMuchandi/rad-skills --skill agent_engine_python_deployer

Installs into the current project.

Summary

Deploys ADK agents to Vertex AI Agent Engine using the Python SDK and registers them with Gemini Enterprise.

Files · 1

View on GitHub

SKILL.md

1	# Agent Engine Python Deployer Skill
2
3	This skill documents the Python SDK method for deploying ADK agents to Vertex AI Agent Engine (`client.agent_engines.create`) and registering them with Gemini Enterprise using the `DiscoveryEngine` HTTP API. This is an alternative to the Terraform-based deployment workflow.
4
5	## 1. Prerequisites
6
7	- Python 3.10+: Active environment with dependencies.
8	- Google Cloud Project: With billing enabled.
9	- Dependencies:
10	```text
11	google-cloud-aiplatform[agent_engines,adk]
12	a2a-sdk >= 0.3.4
13	cloudpickle >= 3.1.2
14	pydantic
15	python-dotenv
16	```
17
18	## 2. Configuration (`.env`)
19
20	Ensure your environment variables are configured correctly:
21
22	```bash
23	# GCP Project Config
24	PROJECT_ID="your-project-id"
25	LOCATION="us-central1"
26	STORAGE_BUCKET="gs://your-staging-bucket"
27
28	# Gemini Enterprise Config
29	GEMINI_ENTERPRISE_APP_ID="your-app-id"
30
31	# Authorization Config (obtained from Discovery Engine authorizations API if needed)
32	AGENT_AUTHORIZATION="projects/PROJECT_NUMBER/locations/global/authorizations/AUTH_ID"
33
34	# Logging & Telemetry flags
35	GOOGLE_CLOUD_AGENT_ENGINE_ENABLE_TELEMETRY=true
36	OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT=true
37	```
38
39	## 3. Core Workflow
40
41	The workflow involves using the Python SDK to deploy and then post-processing the endpoint for Gemini Enterprise.
42
43	### 2.B Create Authorization Resource (If Required)
44
45	When deploying A2A agents (including A2UI and non-A2UI A2A agents on Agent Engine), you must create an `AGENT_AUTHORIZATION` resource in Gemini Enterprise.
46
47	#### 2.B.1 Prerequisite: Create Web OAuth Client
48
49	Before creating the authorization resource, you must create a Web OAuth Client in the Google Cloud Console for the project.
50
51	1. Navigate to APIs & Services > Credentials in the Cloud Console.
52	2. Click Create Credentials > OAuth client ID.
53	3. Select Application type: Web application.
54	4. Add the following Authorized redirect URIs:
55	* `https://vertexaisearch.cloud.google.com/oauth-redirect`
56	* `https://vertexaisearch.cloud.google.com/static/oauth/oauth.html`
57	5. Save the Client ID and Client Secret.
58
59	#### 2.B.2 Construct authorizationUri
60
61	Construct the `authorizationUri` using the Client ID and the encoded redirect URI:
62
63	```text
64	https://accounts.google.com/o/oauth2/v2/auth?client_id=<YOUR_CLIENT_ID>&redirect_uri=https%3A%2F%2Fvertexaisearch.cloud.google.com%2Fstatic%2Foauth%2Foauth.html&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fcloud-platform&include_granted_scopes=true&response_type=code&access_type=offline&prompt=consent
65	```
66
67	#### 2.B.3 Create Authorization Resource via cURL
68
69	```bash
70	curl -X POST \
71	-H "Authorization: Bearer \$(gcloud auth application-default print-access-token)" \
72	-H "Content-Type: application/json" \
73	-H "X-Goog-User-Project: <PROJECT_ID>" \
74	"https://global-discoveryengine.googleapis.com/v1alpha/projects/<PROJECT_ID>/locations/global/authorizations?authorizationId=<AUTH_ID>" \
75	-d '{
76	"name": "projects/<PROJECT_ID>/locations/global/authorizations/<AUTH_ID>",
77	"serverSideOauth2": {
78	"clientId": "<CLIENT_ID>",
79	"clientSecret": "<CLIENT_SECRET>",
80	"authorizationUri": "<AUTH_URI>",
81	"tokenUri": "<TOKEN_URI>"
82	}
83	}'
84	```
85
86	Replace `<PROJECT_ID>`, `<AUTH_ID>`, `<CLIENT_ID>`, `<CLIENT_SECRET>`, `<AUTH_URI>`, and `<TOKEN_URI>`.
87
88	### 3.A Initialize Vertex AI
89	```python
90	import os
91	import vertexai
92	from dotenv import load_dotenv
93
94	load_dotenv()
95
96	vertexai.init(
97	project=os.environ["PROJECT_ID"],
98	location=os.environ["LOCATION"],
99	staging_bucket=os.environ["STORAGE_BUCKET"],
100	)
101	```
102
103	### B. Define `AgentSkill` and Create Card
104	```python
105	from a2a.types import AgentSkill
106	from vertexai.preview.reasoning_engines.templates.a2a import create_agent_card
107
108	agent_skill = AgentSkill(
109	id="contact_lookup",
110	name="Contact Lookup Agent",
111	description="A helpful assistant agent that can find contact card.",
112	tags=["Contact-Lookup"],
113	examples=["Find John Doe"],
114	)
115
116	agent_card = create_agent_card(
117	agent_name="Test Contact Card Agent",
118	description="A helpful assistant agent that can find contact card.",
119	skills=[agent_skill],
120	)
121	```
122
123	### C. Instantiate A2aAgent with Custom Executor
124	Override the default executor to intercept the stream and parse A2UI payloads, or simply yield standard A2A messages for non-A2UI agents.
125	```python
126	from vertexai.preview.reasoning_engines import A2aAgent
127	import agent_executor # Your custom executor file
128
129	a2a_agent = A2aAgent(
130	agent_card=agent_card,
131	agent_executor_builder=agent_executor.AdkAgentToA2AExecutor,
132	)
133	```
134
135	### D. Deploy New Instance with Extra Packages
136	Use `extra_packages` to bundle non-module dependencies (like external scripts or templates).
137	```python
138	from google.genai import types
139
140	client = vertexai.Client(
141	project=o

People who liked this also viewed

6 picks

Type

microsoft-foundryDeploy, evaluate, fine-tune, and manage Foundry agents end-to-end with azd: hosted agent scaffold/run/deploy, prompt agent create, batch eval, continuous eval, prompt optimizer, Agent Optimizer scaffoSkillsJul 2026476k1.3k
azure-deployExecute Azure deployments for ALREADY-PREPARED applications that have existing .azure/deployment-plan.md and infrastructure files. DO NOT use this skill when the user asks to CREATE a new application SkillsJul 2026472k1.3k
azure-preparePrepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wSkillsJul 2026471k1.3k
azure-validate--- name: azure-validate description: "Pre-deployment validation for Azure readiness. Run deep checks on configuration, infrastructure (Bicep or Terraform), RBAC role assignments, managed identity pSkillsJul 2026471k1.3k
azure-aigatewayConfigure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, jailbSkillsJul 2026470k1.3k
azure-hosted-copilot-sdkOfficial agent plugin providing skills and MCP server configurations for Azure scenarios.SkillsJul 2026424k1.3k

agent_engine_python_deployer

15stars1forks

View on GitHub

How to install agent_engine_python_deployer?

veermuchandi/rad-skills/agent_engine_python_deployer

$npx -y skills add VeerMuchandi/rad-skills --skill agent_engine_python_deployer

Installs into the current project.

Summary

Deploys ADK agents to Vertex AI Agent Engine using the Python SDK and registers them with Gemini Enterprise.

Files · 1

View on GitHub

SKILL.md

1	# Agent Engine Python Deployer Skill
2
3	This skill documents the Python SDK method for deploying ADK agents to Vertex AI Agent Engine (`client.agent_engines.create`) and registering them with Gemini Enterprise using the `DiscoveryEngine` HTTP API. This is an alternative to the Terraform-based deployment workflow.
4
5	## 1. Prerequisites
6
7	- Python 3.10+: Active environment with dependencies.
8	- Google Cloud Project: With billing enabled.
9	- Dependencies:
10	```text
11	google-cloud-aiplatform[agent_engines,adk]
12	a2a-sdk >= 0.3.4
13	cloudpickle >= 3.1.2
14	pydantic
15	python-dotenv
16	```
17
18	## 2. Configuration (`.env`)
19
20	Ensure your environment variables are configured correctly:
21
22	```bash
23	# GCP Project Config
24	PROJECT_ID="your-project-id"
25	LOCATION="us-central1"
26	STORAGE_BUCKET="gs://your-staging-bucket"
27
28	# Gemini Enterprise Config
29	GEMINI_ENTERPRISE_APP_ID="your-app-id"
30
31	# Authorization Config (obtained from Discovery Engine authorizations API if needed)
32	AGENT_AUTHORIZATION="projects/PROJECT_NUMBER/locations/global/authorizations/AUTH_ID"
33
34	# Logging & Telemetry flags
35	GOOGLE_CLOUD_AGENT_ENGINE_ENABLE_TELEMETRY=true
36	OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT=true
37	```
38
39	## 3. Core Workflow
40
41	The workflow involves using the Python SDK to deploy and then post-processing the endpoint for Gemini Enterprise.
42
43	### 2.B Create Authorization Resource (If Required)
44
45	When deploying A2A agents (including A2UI and non-A2UI A2A agents on Agent Engine), you must create an `AGENT_AUTHORIZATION` resource in Gemini Enterprise.
46
47	#### 2.B.1 Prerequisite: Create Web OAuth Client
48
49	Before creating the authorization resource, you must create a Web OAuth Client in the Google Cloud Console for the project.
50
51	1. Navigate to APIs & Services > Credentials in the Cloud Console.
52	2. Click Create Credentials > OAuth client ID.
53	3. Select Application type: Web application.
54	4. Add the following Authorized redirect URIs:
55	* `https://vertexaisearch.cloud.google.com/oauth-redirect`
56	* `https://vertexaisearch.cloud.google.com/static/oauth/oauth.html`
57	5. Save the Client ID and Client Secret.
58
59	#### 2.B.2 Construct authorizationUri
60
61	Construct the `authorizationUri` using the Client ID and the encoded redirect URI:
62
63	```text
64	https://accounts.google.com/o/oauth2/v2/auth?client_id=<YOUR_CLIENT_ID>&redirect_uri=https%3A%2F%2Fvertexaisearch.cloud.google.com%2Fstatic%2Foauth%2Foauth.html&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fcloud-platform&include_granted_scopes=true&response_type=code&access_type=offline&prompt=consent
65	```
66
67	#### 2.B.3 Create Authorization Resource via cURL
68
69	```bash
70	curl -X POST \
71	-H "Authorization: Bearer \$(gcloud auth application-default print-access-token)" \
72	-H "Content-Type: application/json" \
73	-H "X-Goog-User-Project: <PROJECT_ID>" \
74	"https://global-discoveryengine.googleapis.com/v1alpha/projects/<PROJECT_ID>/locations/global/authorizations?authorizationId=<AUTH_ID>" \
75	-d '{
76	"name": "projects/<PROJECT_ID>/locations/global/authorizations/<AUTH_ID>",
77	"serverSideOauth2": {
78	"clientId": "<CLIENT_ID>",
79	"clientSecret": "<CLIENT_SECRET>",
80	"authorizationUri": "<AUTH_URI>",
81	"tokenUri": "<TOKEN_URI>"
82	}
83	}'
84	```
85
86	Replace `<PROJECT_ID>`, `<AUTH_ID>`, `<CLIENT_ID>`, `<CLIENT_SECRET>`, `<AUTH_URI>`, and `<TOKEN_URI>`.
87
88	### 3.A Initialize Vertex AI
89	```python
90	import os
91	import vertexai
92	from dotenv import load_dotenv
93
94	load_dotenv()
95
96	vertexai.init(
97	project=os.environ["PROJECT_ID"],
98	location=os.environ["LOCATION"],
99	staging_bucket=os.environ["STORAGE_BUCKET"],
100	)
101	```
102
103	### B. Define `AgentSkill` and Create Card
104	```python
105	from a2a.types import AgentSkill
106	from vertexai.preview.reasoning_engines.templates.a2a import create_agent_card
107
108	agent_skill = AgentSkill(
109	id="contact_lookup",
110	name="Contact Lookup Agent",
111	description="A helpful assistant agent that can find contact card.",
112	tags=["Contact-Lookup"],
113	examples=["Find John Doe"],
114	)
115
116	agent_card = create_agent_card(
117	agent_name="Test Contact Card Agent",
118	description="A helpful assistant agent that can find contact card.",
119	skills=[agent_skill],
120	)
121	```
122
123	### C. Instantiate A2aAgent with Custom Executor
124	Override the default executor to intercept the stream and parse A2UI payloads, or simply yield standard A2A messages for non-A2UI agents.
125	```python
126	from vertexai.preview.reasoning_engines import A2aAgent
127	import agent_executor # Your custom executor file
128
129	a2a_agent = A2aAgent(
130	agent_card=agent_card,
131	agent_executor_builder=agent_executor.AdkAgentToA2AExecutor,
132	)
133	```
134
135	### D. Deploy New Instance with Extra Packages
136	Use `extra_packages` to bundle non-module dependencies (like external scripts or templates).
137	```python
138	from google.genai import types
139
140	client = vertexai.Client(
141	project=o

People who liked this also viewed

6 picks

Type