CLI application for working with ChatGPT. Interactive or file based session with support for a directory of prompts.
I built this for 2 reasons
- To use in vim with
!!
and stdin/out - To do prompt engineering and save sessions
go install github.com/verdverm/chatgpt@latest
chatgpt -h
Set OPENAI_API_KEY
, which you can get here: https://platform.openai.com/account/api-keys
Work with ChatGPT in console.
Examples:
# start an interactive session with gpt-3.5 or gpt-4
chatgpt -i
# ask chatgpt for a one-time response
chatgpt -q "answer me this ChatGPT..."
# provide context to a question or conversation
chatgpt context.txt -i
chatgpt context.txt -q "answer me this ChatGPT..."
# read prompt from file and --write response back
chatgpt convo.txt
chatgpt convo.txt --write
# pipe content from another program, useful for ! in vim visual mode
cat convo.txt | chatgpt
# set the directory for custom prompts, defaults to "prompts"
chatgpt -P custom-prompts -p my-prompt -i
# inspect the predifined prompts, which set ChatGPT's mood
chatgpt -p list
chatgpt -p view:<name>
# use a prompts with any of the previous modes
chatgpt -p optimistic -i
chatgpt -p cynic -q "Is the world going to be ok?"
chatgpt -p teacher convo.txt
# edit mode
chatgpt -e ...
# code mode
chatgpt -c ...
# model options (https://platform.openai.com/docs/api-reference/completions/create)
chatgpt -T 4096 # set max tokens in reponse [0,4096]
chatgpt -C # clean whitespace before sending
chatgpt -E # echo back the prompt, useful for vim coding
chatgpt --temp # set the temperature param [0.0,2.0]
chatgpt --topp # set the TopP param [0.0,1.0]
chatgpt --pres # set the Presence Penalty [-2.0,2.0]
chatgpt --freq # set the Frequency Penalty [-2.0,2.0]
# change model selection, available models are listed here:
# https://pkg.go.dev/github.com/sashabaranov/go-openai#Client.ListModels
chatgpt -m text-davinci-003 # set the model to text-davinci-003 (the default)
chatgpt -m text-ada-001 # set the model to text-ada-001
Usage:
chatgpt [file] [flags]
Flags:
-x, --clean remove excess whitespace from prompt before sending
-C, --count int set the number of response options to create (default 1)
-E, --echo Echo back the prompt, useful for vim coding
-e, --edit request an edit with ChatGPT
--freq float set the Frequency Penalty parameter
-h, --help help for chatgpt
-i, --interactive start an interactive session with ChatGPT
-m, --model string select the model to use with -q or -e (default "text-davinci-003")
--pres float set the Presence Penalty parameter
-p, --prompt string prompt to add to ChatGPT input, use 'list' or 'view:<name>' to inspect predefined, '<name>' to use a prompt, or otherwise supply any custom text
-P, --prompt-dir string directory containing custom prompts, if not set the embedded defaults are used (default "prompts")
-q, --question string ask a single question and print the response back
--temp float set the temperature parameter (default 0.7)
-T, --tokens int set the MaxTokens to generate per response (default 1024)
--topp float set the TopP parameter (default 1)
--version print version information
-w, --write write response to end of context file
Set CHATGPT_PROMPT_DIR
to set the default for -P/--prompt-dir
$ chatgpt -p list -P ./prompts
coding
cynic
liar
optimistic
sam
teacher
thoughtful
$ chatgpt -i
using chat compatible model: gpt-3.5-turbo
starting interactive session...
'quit' to exit
'save <filename>' to preserve
'clear' to erase the context
'context' to see the current context
'prompt' to set the context to a prompt
'tokens' to change the MaxToken param
'count' to change number of responses
'temp' set the temperature param [0.0,2.0]
'topp' set the TopP param [0.0,1.0]
'pres' set the Presence Penalty [-2.0,2.0]
'freq' set the Frequency Penalty [-2.0,2.0]
'model' to change the selected model
>
- https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/
- https://learnprompting.org/
- https://github.com/dair-ai/Prompt-Engineering-Guide
- https://github.com/promptslab/Awesome-Prompt-Engineering
- https://old.reddit.com/r/ChatGPT/comments/10tevu1/new_jailbreak_proudly_unveiling_the_tried_and/
Happily accepting interesting prompts, code improvements, or anything else