a small article on AI in CLIs

pulp_docs/blog/posts/2025/ai_cli_integration.md

@@ -0,0 +1,108 @@
+---
+date: 2025-06-17T00:00:00Z
+title: AI CLI Integration - An experiment
+author: Yasen Trahnov
+tags:
+  - documentation
+  - AI
+---
+# What Happens When You Add AI to Your Terminal? A Personal Experiment
+
+## A Simple Idea
+
+I wanted to try adding AI to a CLI tool to see how easy it would be. The idea was simple: instead of looking up command syntax, just ask the terminal in plain English.
+
+<!-- more -->
+
+## Building It
+
+Here's the basic implementation using OpenAI's API, which than got ported to pulp-cli:
+
+```python
+import json
+import click
+import requests
+
+@click.command()
+@click.argument("question", type=str)
+@click.option("--api-key", envvar="OPENAI_API_KEY", help="OpenAI API key")
+@click.option("--model", default="gpt-3.5-turbo", help="ChatGPT model to use")
+def ask(question, api_key, model):
+    """Ask a question to ChatGPT and get a response."""
+
+    if not api_key:
+        raise click.ClickException("OpenAI API key is required")
+
+    url = "https://api.openai.com/v1/chat/completions"
+    headers = {
+        "Authorization": f"Bearer {api_key}",
+        "Content-Type": "application/json",
+    }
+
+    payload = {
+        "model": model,
+        "messages": [
+            {"role": "user", "content": question}
+        ],
+        "max_tokens": 150,
+        "temperature": 0.7,
+    }
+
+    try:
+        response = requests.post(url, headers=headers, json=payload, timeout=30)
+        response.raise_for_status()
+
+        result = response.json()
+        answer = result["choices"][0]["message"]["content"].strip()
+        click.echo(answer)
+
+    except requests.exceptions.RequestException as e:
+        raise click.ClickException(f"API request failed: {str(e)}")
+    except (json.JSONDecodeError, KeyError):
+        raise click.ClickException("Failed to parse API response")
+
+if __name__ == "__main__":
+    ask()
+```
+
+## Usage Examples
+
+```bash
+# Set your API key
+export OPENAI_API_KEY="your-api-key-here"
+
+# Ask questions
+python3 openai.py "how do I list files in linux"
+python3 openai.py "explain git rebase"
+
+# Or pass the API key directly
+python3 openai.py --api-key "your-key" "what is docker"
+```
+
+For Pulp CLI integration:
+```bash
+# Ask about repository management
+pulp console chat ask "How do I create an RPM repository?"
+
+# Get help with workflows
+pulp console chat ask --model gpt-4.1 "How do I sync and publish content?"
+
+# Troubleshooting
+pulp console chat ask "Why is my sync failing with authentication errors?"
+```
+
+## How Easy It Is
+
+It's straightforward to execute, using the OpenAI documentation. The actual implementation includes additional features like:
+
+- Multiple output formats (JSON/text)
+- Model selection options  
+- Configurable response parameters
+- Better error handling
+
+You can add this to any CLI tool by:
+1. Adding the chat command to your CLI framework
+2. Setting up the OpenAI API integration
+3. Handling API keys and error cases
+
+The full implementation is available in the [Pulp CLI Console extension](https://github.com/pulp/pulp-cli-console) - check out <https://github.com/pulp/pulp-cli-console/blob/pulp-cli-ai-experiment/pulpcore/cli/console/chat.py>.