Switch to ChatGPT API by default, allow other models (#27)

* Switch to ChatGPT API by default, allow other models

aiac will now use the ChatGPT API (and thus the gpt-3.5-turbo
model) instead of the text-davinci-003 model by default. It also
allows selecting the model to use, with the aforementioned two
being supported, along with the code-davinci-002 model, which is
specifically designed to generate code. A model can be selected
via the `--model` flag, and a list of all supported models is
available via the `list-models` command.

Since ChatGPT usually returns a Markdown-formatted description
rather than just code, the library will extract the code from the
response. A new flag, `--full`, is added to prevent that and echo
or save the complete response from the API.

The command line prompt will now also allow users to modify the
prompt after receiving a response (using the "m" key), along with
the previous option to retry with the same prompt.

Also included are small fixes: entering Ctrl+D and Ctrl+C after
receiving a response will quit the program as expected, rather than
regenerate a response.

---------

Co-authored-by: Liav Yona <[email protected]>

.golangci.yml

@@ -24,6 +24,9 @@ linters:
     - promlinter
     - testpackage
     - varnamelen
+    - nonamedreturns
+    - tagliatelle
+    - exhaustruct
 issues:
   new: true
   new-from-rev: HEAD

README.md

@@ -20,6 +20,7 @@ Generator.
 * [Instructions](#instructions)
     * [Installation](#installation)
     * [Usage](#usage)
+    * [Choosing a Different Model](#choosing-a-different-model)
 * [Example Output](#example-output)
 * [Troubleshooting](#troubleshooting)
 * [Support Channels](#support-channels)
@@ -33,7 +34,8 @@ Generator.
 via [OpenAI](https://openai.com/)'s API. The CLI allows you to ask the model to generate templates
 for different scenarios (e.g. "get terraform for AWS EC2"). It will make the
 request, and store the resulting code to a file, or simply print it to standard
-output.
+output. By default, `aiac` uses the same model used by ChatGPT, but allows using
+different models.
 
 ## Use Cases and Example Prompts
 
@@ -92,7 +94,7 @@ Using `docker`:
 
 Using `go install`:
 
-    go install github.com/gofireflyio/aiac/v2
+    go install github.com/gofireflyio/aiac/v2@latest
 
 Alternatively, clone the repository and build from source:
 
@@ -101,12 +103,12 @@ Alternatively, clone the repository and build from source:
 
 ### Usage
 
-1. Create your OpenAI API key [here](https://beta.openai.com/account/api-keys).
+1. Create your OpenAI API key [here](platform.openai.com/account/api-keys).
 2. Click “Create new secret key” and copy it.
 3. Provide the API key via the `OPENAI_API_KEY` environment variable or via the `--api-key` command line flag.
 
 By default, aiac prints the extracted code to standard output and asks if it
-should save or regenerate the code:
+should save the code, regenerate it, or modify the prompt:
 
     aiac get terraform for AWS EC2
 
@@ -121,6 +123,15 @@ To run using `docker`:
     -e OPENAI_API_KEY=[PUT YOUR KEY HERE] \
     ghcr.io/gofireflyio/aiac get terraform for ec2
 
+If you want to receive and/or store the complete Markdown output from OpenAI,
+including explanations (if any), use the `--full` flag.
+
+### Choosing a Different Model
+
+Use the `--model` flag to select a different model than the default (currently
+"gpt-3.5-turbo"). Not all OpenAI models are supported, use `aiac list-models`
+to get a list of all supported models.
+
 ## Example Output
 
 Command line prompt: