gemini-api-integration

Name: gemini-api-integration
Author: tranhieutt

69stars41forks

How to install gemini-api-integration?

tranhieutt/software_development_department/gemini-api-integration

$npx -y skills add tranhieutt/software_development_department --skill gemini-api-integration

Installs into the current project.

Summary

Provides code patterns for Google Gemini API integration including text generation, multimodal inputs, and streaming. Use when working with Google AI SDK or when the user mentions Gemini API, Google AI, or Vertex AI.

Files · 1

View on GitHub

SKILL.md

1	# Gemini API Integration
2
3	## Overview
4
5	This skill guides AI agents through integrating Google Gemini API into applications — from basic text generation to advanced multimodal, function calling, and streaming use cases. It covers the full Gemini SDK lifecycle with production-grade patterns.
6
7	## When to Use This Skill
8
9	- Use when setting up Gemini API for the first time in a Node.js, Python, or browser project
10	- Use when implementing multimodal inputs (text + image/audio/video)
11	- Use when adding streaming responses to improve perceived latency
12	- Use when implementing function calling / tool use with Gemini
13	- Use when optimizing model selection (Flash vs Pro vs Ultra) for cost and performance
14	- Use when debugging Gemini API errors, rate limits, or quota issues
15
16	## Step-by-Step Guide
17
18	### 1. Installation & Setup
19
20	Node.js / TypeScript:
21	```bash
22	npm install @google/generative-ai
23	```
24
25	Python:
26	```bash
27	pip install google-generativeai
28	```
29
30	Set your API key securely:
31	```bash
32	export GEMINI_API_KEY="your-api-key-here"
33	```
34
35	### 2. Basic Text Generation
36
37	Node.js:
38	```javascript
39	import { GoogleGenerativeAI } from "@google/generative-ai";
40
41	const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
42	const model = genAI.getGenerativeModel({ model: "gemini-1.5-flash" });
43
44	const result = await model.generateContent("Explain async/await in JavaScript");
45	console.log(result.response.text());
46	```
47
48	Python:
49	```python
50	import google.generativeai as genai
51	import os
52
53	genai.configure(api_key=os.environ["GEMINI_API_KEY"])
54	model = genai.GenerativeModel("gemini-1.5-flash")
55
56	response = model.generate_content("Explain async/await in JavaScript")
57	print(response.text)
58	```
59
60	### 3. Streaming Responses
61
62	```javascript
63	const result = await model.generateContentStream("Write a detailed blog post about AI");
64
65	for await (const chunk of result.stream) {
66	process.stdout.write(chunk.text());
67	}
68	```
69
70	### 4. Multimodal Input (Text + Image)
71
72	```javascript
73	import fs from "fs";
74
75	const imageData = fs.readFileSync("screenshot.png");
76	const imagePart = {
77	inlineData: {
78	data: imageData.toString("base64"),
79	mimeType: "image/png",
80	},
81	};
82
83	const result = await model.generateContent(["Describe this image:", imagePart]);
84	console.log(result.response.text());
85	```
86
87	### 5. Function Calling / Tool Use
88
89	```javascript
90	const tools = [{
91	functionDeclarations: [{
92	name: "get_weather",
93	description: "Get current weather for a city",
94	parameters: {
95	type: "OBJECT",
96	properties: {
97	city: { type: "STRING", description: "City name" },
98	},
99	required: ["city"],
100	},
101	}],
102	}];
103
104	const model = genAI.getGenerativeModel({ model: "gemini-1.5-pro", tools });
105	const result = await model.generateContent("What's the weather in Mumbai?");
106
107	const call = result.response.functionCalls()?.[0];
108	if (call) {
109	// Execute the actual function
110	const weatherData = await getWeather(call.args.city);
111	// Send result back to model
112	}
113	```
114
115	### 6. Multi-turn Chat
116
117	```javascript
118	const chat = model.startChat({
119	history: [
120	{ role: "user", parts: [{ text: "You are a helpful coding assistant." }] },
121	{ role: "model", parts: [{ text: "Sure! I'm ready to help with code." }] },
122	],
123	});
124
125	const response = await chat.sendMessage("How do I reverse a string in Python?");
126	console.log(response.response.text());
127	```
128
129	### 7. Model Selection Guide
130
131	\| Model \| Best For \| Speed \| Cost \|
132	\|-------\|----------\|-------\|------\|
133	\| `gemini-1.5-flash` \| High-throughput, cost-sensitive tasks \| Fast \| Low \|
134	\| `gemini-1.5-pro` \| Complex reasoning, long context \| Medium \| Medium \|
135	\| `gemini-2.0-flash` \| Latest fast model, multimodal \| Very Fast \| Low \|
136	\| `gemini-2.0-pro` \| Most capable, advanced tasks \| Slow \| High \|
137
138	## Best Practices
139
140	- ✅ Do: Use `gemini-1.5-flash` for most tasks — it's fast and cost-effective
141	- ✅ Do: Always stream responses for user-facing chat UIs to reduce perceived latency
142	- ✅ Do: Store API keys in environment variables, never hard-code them
143	- ✅ Do: Implement exponential backoff for rate limit (429) errors
144	- ✅ Do: Use `systemInstruction` to set persistent model behavior
145	- ❌ Don't: Use `gemini-pro` for simple tasks — Flash is cheaper and faster
146	- ❌ Don't: Send large base64 images inline for files > 20MB — use File API instead
147	- ❌ Don't: Ignore safety ratings in responses for production apps
148
149	## Error Handling
150
151	```javascript
152	try {
153	const result = await mo

People who liked this also viewed

6 picks

Type

gemini-api-integration

69stars41forks

View on GitHub

How to install gemini-api-integration?

tranhieutt/software_development_department/gemini-api-integration

$npx -y skills add tranhieutt/software_development_department --skill gemini-api-integration

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Gemini API Integration
2
3	## Overview
4
5	This skill guides AI agents through integrating Google Gemini API into applications — from basic text generation to advanced multimodal, function calling, and streaming use cases. It covers the full Gemini SDK lifecycle with production-grade patterns.
6
7	## When to Use This Skill
8
9	- Use when setting up Gemini API for the first time in a Node.js, Python, or browser project
10	- Use when implementing multimodal inputs (text + image/audio/video)
11	- Use when adding streaming responses to improve perceived latency
12	- Use when implementing function calling / tool use with Gemini
13	- Use when optimizing model selection (Flash vs Pro vs Ultra) for cost and performance
14	- Use when debugging Gemini API errors, rate limits, or quota issues
15
16	## Step-by-Step Guide
17
18	### 1. Installation & Setup
19
20	Node.js / TypeScript:
21	```bash
22	npm install @google/generative-ai
23	```
24
25	Python:
26	```bash
27	pip install google-generativeai
28	```
29
30	Set your API key securely:
31	```bash
32	export GEMINI_API_KEY="your-api-key-here"
33	```
34
35	### 2. Basic Text Generation
36
37	Node.js:
38	```javascript
39	import { GoogleGenerativeAI } from "@google/generative-ai";
40
41	const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
42	const model = genAI.getGenerativeModel({ model: "gemini-1.5-flash" });
43
44	const result = await model.generateContent("Explain async/await in JavaScript");
45	console.log(result.response.text());
46	```
47
48	Python:
49	```python
50	import google.generativeai as genai
51	import os
52
53	genai.configure(api_key=os.environ["GEMINI_API_KEY"])
54	model = genai.GenerativeModel("gemini-1.5-flash")
55
56	response = model.generate_content("Explain async/await in JavaScript")
57	print(response.text)
58	```
59
60	### 3. Streaming Responses
61
62	```javascript
63	const result = await model.generateContentStream("Write a detailed blog post about AI");
64
65	for await (const chunk of result.stream) {
66	process.stdout.write(chunk.text());
67	}
68	```
69
70	### 4. Multimodal Input (Text + Image)
71
72	```javascript
73	import fs from "fs";
74
75	const imageData = fs.readFileSync("screenshot.png");
76	const imagePart = {
77	inlineData: {
78	data: imageData.toString("base64"),
79	mimeType: "image/png",
80	},
81	};
82
83	const result = await model.generateContent(["Describe this image:", imagePart]);
84	console.log(result.response.text());
85	```
86
87	### 5. Function Calling / Tool Use
88
89	```javascript
90	const tools = [{
91	functionDeclarations: [{
92	name: "get_weather",
93	description: "Get current weather for a city",
94	parameters: {
95	type: "OBJECT",
96	properties: {
97	city: { type: "STRING", description: "City name" },
98	},
99	required: ["city"],
100	},
101	}],
102	}];
103
104	const model = genAI.getGenerativeModel({ model: "gemini-1.5-pro", tools });
105	const result = await model.generateContent("What's the weather in Mumbai?");
106
107	const call = result.response.functionCalls()?.[0];
108	if (call) {
109	// Execute the actual function
110	const weatherData = await getWeather(call.args.city);
111	// Send result back to model
112	}
113	```
114
115	### 6. Multi-turn Chat
116
117	```javascript
118	const chat = model.startChat({
119	history: [
120	{ role: "user", parts: [{ text: "You are a helpful coding assistant." }] },
121	{ role: "model", parts: [{ text: "Sure! I'm ready to help with code." }] },
122	],
123	});
124
125	const response = await chat.sendMessage("How do I reverse a string in Python?");
126	console.log(response.response.text());
127	```
128
129	### 7. Model Selection Guide
130
131	\| Model \| Best For \| Speed \| Cost \|
132	\|-------\|----------\|-------\|------\|
133	\| `gemini-1.5-flash` \| High-throughput, cost-sensitive tasks \| Fast \| Low \|
134	\| `gemini-1.5-pro` \| Complex reasoning, long context \| Medium \| Medium \|
135	\| `gemini-2.0-flash` \| Latest fast model, multimodal \| Very Fast \| Low \|
136	\| `gemini-2.0-pro` \| Most capable, advanced tasks \| Slow \| High \|
137
138	## Best Practices
139
140	- ✅ Do: Use `gemini-1.5-flash` for most tasks — it's fast and cost-effective
141	- ✅ Do: Always stream responses for user-facing chat UIs to reduce perceived latency
142	- ✅ Do: Store API keys in environment variables, never hard-code them
143	- ✅ Do: Implement exponential backoff for rate limit (429) errors
144	- ✅ Do: Use `systemInstruction` to set persistent model behavior
145	- ❌ Don't: Use `gemini-pro` for simple tasks — Flash is cheaper and faster
146	- ❌ Don't: Send large base64 images inline for files > 20MB — use File API instead
147	- ❌ Don't: Ignore safety ratings in responses for production apps
148
149	## Error Handling
150
151	```javascript
152	try {
153	const result = await mo

People who liked this also viewed

6 picks

Type