feat: Add gemini sample

samples/google/gemini/README.md

@@ -0,0 +1,77 @@
+---
+title: Gemini
+description: Simple usage of the Gemini API
+integrations: ["gemini"]
+categories: ["Samples"]
+---
+
+# Gemini Sample
+
+This [AutoKitteh](https://github.com/autokitteh/autokitteh) project
+demonstrates integration with [Gemini](https://gemini.google.com).
+
+It sends a couple of requests to the Gemini API, and prints the responses
+back to the user.
+
+API details:
+
+- [Gemini API](https://ai.google.dev)
+- [Python client library](https://github.com/google-gemini/generative-ai-python/blob/main/docs/api/google/generativeai.md)
+
+## How It Works
+
+1. Generate content using Gemini and log the response.
+2. Conduct interactive chats with Gemini and log the responses.
+
+## Deployment & Configuration
+
+### Cloud Usage
+
+- Initialize your connection with Gemini through the UI
+
+### Self-Hosted Server
+
+#### Prerequisites
+
+- [Install AutoKitteh](https://docs.autokitteh.com/get_started/install)
+
+#### Installation Steps
+
+1. Clone the repository:
+   ```shell
+   git clone https://github.com/autokitteh/kittehub.git
+   cd kittehub/samples/google/gemini
+   ```
+
+2. Start the AutoKitteh server:
+   ```shell
+   ak up --mode dev
+   ```
+
+3. Deploy the project:
+   ```shell
+   ak deploy --manifest autokitteh.yaml
+   ```
+
+   The output will show your connection IDs, which you'll need for the next step. Look for lines like:
+   ```shell
+   [exec] create_connection "gemini_sample/gemini_connection": con_01je39d6frfdtshstfg5qpk8sz created
+   ```
+
+   In this example, `con_01je39d6frfdtshstfg5qpk8sz` is the connection ID.
+
+4. Initialize your connections using the CLI:
+   ```shell
+   ak connection init gemini_connection <connection ID>
+   ```
+
+## Trigger Workflow
+
+The workflow is triggered by sending an HTTP GET request.
+
+> [!TIP]
+> The workflow can also be triggered manually by clicking the "Run" button in the UI. Make sure to set the `on_http_get` function as the entrypoint.
+
+> [!IMPORTANT]
+> Ensure that the connection with Gemini is properly initialized before the workflow starts running.
+

samples/google/gemini/autokitteh.yaml

@@ -0,0 +1,15 @@
+# This YAML file is a declarative manifest that describes the setup
+# of an AutoKitteh project that demonstrates integration with Gemini.
+
+version: v1
+
+project:
+  name: gemini_sample
+  connections:
+    - name: gemini_conn
+      integration: googlegemini
+  triggers:
+    - name: on_http_get
+      type: webhook
+      event_type: get
+      call: program.py:on_http_get

samples/google/gemini/program.py

@@ -0,0 +1,34 @@
+"""Demonstration of AutoKitteh's Gemini integration.
+
+A single entry-point function is implemented, it sends a couple of requests to the Gemini API, and prints the responses
+in the AutoKitteh session log.
+"""
+
+from autokitteh.google import gemini_client
+
+MODEL = "gemini-1.5-flash"
+
+gemini = gemini_client("gemini_conn", model_name=MODEL)
+
+
+def on_http_get(event):
+    """Entry-point function for handling HTTP GET requests in this workflow."""
+    # Example 1: trivial interaction with Gemini.
+    prompt = "say meow in different languages"
+    response = gemini.generate_content(prompt)
+    print(response.text)
+
+    # Example 2: interactive chat using the Gemini.
+    chat = gemini.start_chat(
+        history=[
+            {"role": "user", "parts": "Hello"},
+            {
+                "role": "model",
+                "parts": "Great to meet you. What would you like to know?",
+            },
+        ]
+    )
+    response = chat.send_message("I have 2 cats in my house.")
+    print(response.text)
+    response = chat.send_message("How many paws are in my house?")
+    print(response.text)