
Security News
Deno 2.2 Improves Dependency Management and Expands Node.js Compatibility
Deno 2.2 enhances Node.js compatibility, improves dependency management, adds OpenTelemetry support, and expands linting and task automation for developers.
Hello! Welcome to Prapti. It's early, and we're just getting started. Please use GitHub Discussions for general queries and suggestions, and Issues for bug reports. See CONTRIBUTING.md for code contributions.
Prapti is a tool for prompting Large Language Models (LLMs) with an editable history. You work within a markdown file where you can edit your prompts and the LLM's responses. Press a hot-key to get the next response from the LLM. Once received, the response is automatically appended to the file.
Using prapti is an interactive experience, but you can get an idea of what it might be like by reading your first prapti conversation.
Prapti requires Python 3.10 or newer.
Installation involves the following required steps:
prapti
command line toolprapti
tool runs manually in your terminalprapti
in your editorprapti
command line toolIn your terminal, run:
pip install git+https://github.com/prapti-ai/prapti
(or pip3
or py -3 -m pip
depending on your system).
We recommend running prapti in a Python virtual environment such as venv.
export OPENAI_API_KEY=your-key-goes-here
(or setx
, depending on your system. There are other ways to manage your environment variables. For details see OpenAI's Best Practices for API Key Safety ).
To select a non-default organisation, use the OPENAI_ORGANIZATION
environment variable.
To use an alternate OpenAI API key and/or organization specifically with prapti you can set PRAPTI_OPENAI_API_KEY
and/or PRAPTI_OPENAI_ORGANIZATION
. If set, prapti will use these environment variables in preference to OPENAI_API_KEY
and OPENAI_ORGANIZATION
.
Click here for local LLM configuration
prapti
tool runs manually in your terminalFirst, create a new markdown file with .md
extension. You can name it anything you like, for example test-chat.md
.
Edit the file so that the final lines consist of a user prompt. A user prompt is a message that you, as the user, write to the LLM. It should look like this:
### @user:
Write your prompt here.
Then run prapti
in the terminal to generate a new assistant response:
prapti test-chat.md
After running prapti
, refresh or reload the test-chat.md
file. You should see that prapti has appended an assistant response to the file. With this test, you've confirmed that prapti works correctly in your terminal. The next step is to set up a convenient way to run prapti, such as a hot-key, for a smoother conversation flow with the LLM.
prapti
in your text editorChoose a key combination (e.g. Ctrl-Enter
) in your preferred text editor, and configure it to save the current file and run prapti
. This keybinding should be set to function only when you're editing markdown files.
Below are the instructions for VSCode. If you use another editor please contribute instructions.
NOTE: This key binding runs
prapti
in the active VSCode terminal window. So make sure you have the terminal open with theprapti
command available.
First use the Quick Open menu (Cmd-Shift-P on Mac, Ctrl-Shift-P on Windows) to run:
Preferences: Open Keyboard Shortcuts (JSON)
then add the following binding to the opened keybindings.json
file.
{
"key": "ctrl+enter",
"command": "runCommands",
"args":{
"commands":[
"workbench.action.files.save",
"cursorBottom",
{
"command": "workbench.action.terminal.sendSequence",
"args": { "text": "prapti ${file}\u000D" }
},
"cursorBottom",
]
},
"when": "editorLangId == markdown"
},
Now, when editing your markdown file you should be able to hit Ctrl-Enter
to get a response from the LLM. You can watch the terminal window for progress. Be patient, GPT4 can take 30 seconds to generate a full response.
See our documentation for optional text editor setup to streamline your experience.
This project is MIT licensed.
FAQs
Converse with LLMs directly in markdown files
We found that prapti demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Deno 2.2 enhances Node.js compatibility, improves dependency management, adds OpenTelemetry support, and expands linting and task automation for developers.
Security News
React's CRA deprecation announcement sparked community criticism over framework recommendations, leading to quick updates acknowledging build tools like Vite as valid alternatives.
Security News
Ransomware payment rates hit an all-time low in 2024 as law enforcement crackdowns, stronger defenses, and shifting policies make attacks riskier and less profitable.