Mods!

AI for the command line, built for pipelines.

LLM based AI is really good at interpreting the output of commands and returning the results in CLI friendly text formats like Markdown. Mods is a simple tool that makes it super easy to use AI on the command line and in your pipelines. Mods works with OpenAI, Groq, Azure OpenAI, and LocalAI

To get started, install Mods and check out some of the examples below. Since Mods has built-in Markdown formatting, you may also want to grab Glow to give the output some pizzazz.

What Can It Do?

Mods works by reading standard in and prefacing it with a prompt supplied in the mods arguments. It sends the input text to an LLM and prints out the result, optionally asking the LLM to format the response as Markdown. This gives you a way to "question" the output of a command. Mods will also work on standard in or an argument supplied prompt individually.

Be sure to check out the examples and a list of all the features.

Installation

Mods works with OpenAI compatible endpoints. By default, Mods is configured to support OpenAI's official API and a LocalAI installation running on port 8080. You can configure additional endpoints in your settings file by running mods --settings.

OpenAI

Mods uses GPT-4 by default and will fall back to GPT-3.5 Turbo if it's not available. Set the OPENAI_API_KEY environment variable to a valid OpenAI key, which you can get from here.

Mods can also use the Azure OpenAI service. Set the AZURE_OPENAI_KEY environment variable and configure your Azure endpoint with mods --settings.

LocalAI

LocalAI allows you to run a multitude of models locally. Mods works with the GPT4ALL-J model as setup in this tutorial. You can define more LocalAI models and endpoints with mods --settings.

Groq

Groq provides some models powered by their LPU inference engine. Mods will work with both their models (mixtral-8x7b-32768 and llama2-70b-4096). Set the GROQ_API_KEY environment variable to a valid key, which you can get from here.

Install Mods

# macOS or Linux
brew install charmbracelet/tap/mods

# Windows (with Winget)
winget install mods

# Windows (with Scoop)
scoop bucket add charm https://github.com/charmbracelet/scoop-bucket.git
scoop install mods

# Arch Linux (btw)
yay -S mods

# Debian/Ubuntu
sudo mkdir -p /etc/apt/keyrings
curl -fsSL https://repo.charm.sh/apt/gpg.key | sudo gpg --dearmor -o /etc/apt/keyrings/charm.gpg
echo "deb [signed-by=/etc/apt/keyrings/charm.gpg] https://repo.charm.sh/apt/ * *" | sudo tee /etc/apt/sources.list.d/charm.list
sudo apt update && sudo apt install mods

# Fedora/RHEL
echo '[charm]
name=Charm
baseurl=https://repo.charm.sh/yum/
enabled=1
gpgcheck=1
gpgkey=https://repo.charm.sh/yum/gpg.key' | sudo tee /etc/yum.repos.d/charm.repo
sudo yum install mods

Or, download it:

Packages are available in Debian and RPM formats
Binaries are available for Linux, macOS, and Windows

Or, just install it with go:

go install github.com/charmbracelet/mods@latest

Saving conversations

Conversations save automatically. They are identified by their latest prompt. Similar to Git, conversations have a SHA-1 identifier and a title. Conversations can be updated, maintaining their SHA-1 identifier but changing their title.

Check the features document for more details.

Settings

--settings

Mods lets you tune your query with a variety of settings. You can configure Mods with mods --settings or pass the settings as environment variables and flags.

Dirs

--dirs

Prints the local directories used by Mods to store its data. Useful if you want to back your conversations up, for example.

Model

-m, --model, MODS_MODEL

Mods uses gpt-4 with OpenAI by default, but you can specify any model as long as your account has access to it or you have installed locally with LocalAI.

You can add new models to the settings with mods --settings. You can also specify a model and an API endpoint with -m and -a to use models not in the settings file.

Ask Model

-M --ask-model

Ask which model to use with an interactive prompt.

Title

-t, --title

Set a custom save title for the conversation.

Continue last

-C, --continue-last

Continues the previous conversation.

Continue

-c, --continue

Continue from the last response or a given title or SHA1.

List

-l, --list

Lists all saved conversations.

Show last

-S, --show-last

Show the previous conversation.

Show

-s, --show

Show the saved conversation the given title or SHA1.

Delete

--delete

Deletes the saved conversation with the given title or SHA1.

--delete-older-than=duration

Delete conversations older than the given duration (e.g. 10d, 3w, 1mo, 1y).

If the terminal is interactive, it'll first list the conversations to be deleted and then will ask for confirmation.

If the terminal is not interactive, or if --quiet is provided, it'll delete the conversations without any confirmation.

Format

-f, --format, MODS_FORMAT

Ask the LLM to format the response in a given format. You can edit the text passed to the LLM with mods --settings then changing the format-text value. You'll likely want to use this in with --format-as.

Format As

--format-as, MODS_FORMAT_AS

When --format is on, instructs the LLM about which format you want the output to be. This can be customized with mods --settings.

Role

--role, MODS_ROLE

You can have customized roles in your settings file, which will be fed to the LLM as system messages in order to change its behavior. The --role flag allows you to change which of these custom roles to use.

Raw

-r, --raw, MODS_RAW

Print the raw response without syntax highlighting, even when connect to a TTY.

Max Tokens

--max-tokens, MODS_MAX_TOKENS

Max tokens tells the LLM to respond in less than this number of tokens. LLMs are better at longer responses so values larger than 256 tend to work best.

Temperature

--temp, MODS_TEMP

Sampling temperature is a number between 0.0 and 2.0 and determines how confident the model is in its choices. Higher values make the output more random and lower values make it more deterministic.

Stop

--stop, MODS_STOP

Up to 4 sequences where the API will stop generating further tokens.

Top P

--topp, MODS_TOPP

Top P is an alternative to sampling temperature. It's a number between 0.0 and 2.0 with smaller numbers narrowing the domain from which the model will create its response.

No Limit

--no-limit, MODS_NO_LIMIT

By default, Mods attempts to size the input to the maximum size the allowed by the model. You can potentially squeeze a few more tokens into the input by setting this but also risk getting a max token exceeded error from the OpenAI API.

Include Prompt

-P, --prompt, MODS_INCLUDE_PROMPT

Include prompt will preface the response with the entire prompt, both standard in and the prompt supplied by the arguments.

Include Prompt Args

-p, --prompt-args, MODS_INCLUDE_PROMPT_ARGS

Include prompt args will include only the prompt supplied by the arguments. This can be useful if your standard in content is long and you just a want a summary before the response.

Max Retries

--max-retries, MODS_MAX_RETRIES

The maximum number of retries to failed API calls. The retries happen with an exponential backoff.

Fanciness

--fanciness, MODS_FANCINESS

Your desired level of fanciness.

Quiet

-q, --quiet, MODS_QUIET

Only output errors to standard err. Hides the spinner and success messages that would go to standard err.

Reset Settings

--reset-settings

Backup your old settings file and reset everything to the defaults.

No Cache

--no-cache, MODS_NO_CACHE

Disables conversation saving.

Wrap Words

--word-wrap, MODS_WORD_WRAP

Wrap formatted output at specific width (default is 80)

HTTP Proxy

-x, --http-proxy, MODS_HTTP_PROXY

Use the HTTP proxy to the connect the API endpoints.

Defining custom roles

Roles allow you to set system prompts, for instance, you might want a shell role that instructs the LLM to output shell commands instead of explanations.

In mods, roles are defined in the settings file¹:

roles:
  "shell":
    - you are a shell expert
    - you do not explain anything
    - you simply output one liners to solve the problems you're asked
    - you do not provide any explanation whatsoever, ONLY the command

You can then use said role as:

mods --role shell list files in the current directory

You can verify that the system prompts were set by running mods -S.

Roles can also be loaded from URIs, e.g.:

roles:
  "create_summary":
    - https://raw.githubusercontent.com/danielmiessler/fabric/main/patterns/create_summary/system.md
  "create_summary_2":
    - file:///fabric/patterns/create_summary/system.md

Using within Vim/neovim

You can use mods as an assistant inside Vim. Here are some examples:

:'<,'>w !mods explain this
:'<,'>.!mods improve this code

You can also add user commands for common actions, for example:

command! -range -nargs=0 ModsExplain :'<,'>w !mods explain this, be very succint
command! -range -nargs=* ModsRefactor :'<,'>!mods refactor this to improve its readability
command! -range -nargs=+ Mods :'<,'>w !mods <q-args>

This allows you to visual select some test, and run :ModsExplain, :ModsRefactor, and :Mods your prompt.

Whatcha Think?

We’d love to hear your thoughts on this project. Feel free to drop us a note.

License

MIT

Part of Charm.

Charm热爱开源 • Charm loves open source

You can edit it by running mods --settings. ↩