## I.S.A.A.C - Intelligent System for Advanced Assistance and Companionship
I.S.A.A.C is a completely local, on-terminal AI assistant that lets you use ChatGPT-like features on the terminal so you don't have to switch windows every 2 minutes, it comes with a set of commands and features that you can turn on or off to get the most out of it, I.S.A.A.C can talk to you using locally run speech-to-text and text-to-speech models allowing you to put your fingers to better use.
- Run `pip install py-isaac`.
- Run `isaac`.
- Type `:commands` to list all available commands.
- Type `hello` and you will be prompted to choose a language model provider.
- Select a provider from [Gemini](https://gemini.google.com/) and [Groq](https://console.groq.com/).
- Now, you will be prompted for an API key.
- Create a [Gemini API key](https://ai.google.dev/gemini-api/docs/api-key) if you selected Gemini or a [Groq API key](https://console.groq.com/keys) if you selected Groq.
- Paste the API key to the prompt.
- Done.
### Available commands
- `:toggle`- toggles features on/off.
- - `:toggle speech` to toggle the assistant's speech.
- - `:toggle context` to toggle the use of conversation history for coherent responses.
- - `:toggle hearing` to toggle the assistant's ability to hear you.
`NOTE: for interacting with the assistant only using your voice, turn on both speech and hearing.`
---
- `:select` - selects from available options.
- - `:select lm_provider` to select the language model provider.
- - `:select lm` to select the model for generating responses.
- - `:select voice` to select a [Piper](https://github.com/rhasspy/piper) text-to-speech model for the assistant to speak with.
- - `:select whisper` to select a [Whisper](https://github.com/openai/whisper) speech-to-text model for the assistant to interpret your voice with.
---
- `:key` sets the LLM API key for the selected provider, run this when the assistant can't process your queries, it means the key most proabably expired.
- `:instruct` instructs the model to behave a certain way, using the [system message](https://promptmetheus.com/resources/llm-knowledge-base/system-message).
- `:status` to see status, selected settings and resource consumption.
- `:mute` to mute the assistant while it's speaking.
- `:cmd` to launch a shell session to run shell commands, run the `exit` command in the shell session to get back to the assistant.
- `:commands` to see all available commands.
- `:clear` to clear the terminal.
- `:exit` to turn the assistant off.
---
### Tasks
- [ ] add shell command execution using shebang, e.g. "!ls".
- [ ] add ollama support, user should be able to choose locally running ollama API as language model provider.
- [ ] User profiling, store general likes and dislikes, e.g. "concise answers", "a project they're working on" etc.
### Contributing
If you don't know where to start, pick one of the pending tasks and start with that. If if you want to fix a bug or add other enhancements, please raise an issue first.
Raw data
{
"_id": null,
"home_page": "https://github.com/n1teshy/py-isaac",
"name": "py-isaac",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "AI assistant, local assistant",
"author": "Nitesh Yadav",
"author_email": "nitesh.txt@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/5a/dc/8935d53fec74cc2497463f6547acb84b57db1c6d5bd06c70afb9fcc4d158/py-isaac-0.1.3.tar.gz",
"platform": "Posix; Windows",
"description": "## I.S.A.A.C - Intelligent System for Advanced Assistance and Companionship\r\n\r\n\r\n\r\nI.S.A.A.C is a completely local, on-terminal AI assistant that lets you use ChatGPT-like features on the terminal so you don't have to switch windows every 2 minutes, it comes with a set of commands and features that you can turn on or off to get the most out of it, I.S.A.A.C can talk to you using locally run speech-to-text and text-to-speech models allowing you to put your fingers to better use.\r\n\r\n- Run `pip install py-isaac`.\r\n- Run `isaac`.\r\n- Type `:commands` to list all available commands.\r\n- Type `hello` and you will be prompted to choose a language model provider.\r\n- Select a provider from [Gemini](https://gemini.google.com/) and [Groq](https://console.groq.com/).\r\n- Now, you will be prompted for an API key.\r\n- Create a [Gemini API key](https://ai.google.dev/gemini-api/docs/api-key) if you selected Gemini or a [Groq API key](https://console.groq.com/keys) if you selected Groq.\r\n- Paste the API key to the prompt.\r\n- Done.\r\n\r\n\r\n### Available commands\r\n- `:toggle`- toggles features on/off.\r\n- - `:toggle speech` to toggle the assistant's speech.\r\n- - `:toggle context` to toggle the use of conversation history for coherent responses.\r\n- - `:toggle hearing` to toggle the assistant's ability to hear you.\r\n\r\n`NOTE: for interacting with the assistant only using your voice, turn on both speech and hearing.`\r\n\r\n---\r\n\r\n- `:select` - selects from available options.\r\n- - `:select lm_provider` to select the language model provider.\r\n- - `:select lm` to select the model for generating responses.\r\n- - `:select voice` to select a [Piper](https://github.com/rhasspy/piper) text-to-speech model for the assistant to speak with.\r\n- - `:select whisper` to select a [Whisper](https://github.com/openai/whisper) speech-to-text model for the assistant to interpret your voice with.\r\n---\r\n- `:key` sets the LLM API key for the selected provider, run this when the assistant can't process your queries, it means the key most proabably expired.\r\n- `:instruct` instructs the model to behave a certain way, using the [system message](https://promptmetheus.com/resources/llm-knowledge-base/system-message).\r\n- `:status` to see status, selected settings and resource consumption.\r\n- `:mute` to mute the assistant while it's speaking.\r\n- `:cmd` to launch a shell session to run shell commands, run the `exit` command in the shell session to get back to the assistant.\r\n- `:commands` to see all available commands.\r\n- `:clear` to clear the terminal.\r\n- `:exit` to turn the assistant off.\r\n\r\n---\r\n### Tasks\r\n- [ ] add shell command execution using shebang, e.g. \"!ls\".\r\n- [ ] add ollama support, user should be able to choose locally running ollama API as language model provider.\r\n- [ ] User profiling, store general likes and dislikes, e.g. \"concise answers\", \"a project they're working on\" etc.\r\n\r\n### Contributing\r\nIf you don't know where to start, pick one of the pending tasks and start with that. If if you want to fix a bug or add other enhancements, please raise an issue first.\r\n",
"bugtrack_url": null,
"license": null,
"summary": "Cooles locally run AI assistant",
"version": "0.1.3",
"project_urls": {
"Homepage": "https://github.com/n1teshy/py-isaac"
},
"split_keywords": [
"ai assistant",
" local assistant"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "4a183635f19ebe13e353fd0688884b7affacbcfd2e6fcfc4d66d773314d17ef1",
"md5": "c371c067bfac0a3f03668fac74c7fcf6",
"sha256": "5e566960c3e543b021a378a6bf660d04e007c28148138814f591849be66bad0e"
},
"downloads": -1,
"filename": "py_isaac-0.1.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "c371c067bfac0a3f03668fac74c7fcf6",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 29594,
"upload_time": "2025-02-15T01:09:46",
"upload_time_iso_8601": "2025-02-15T01:09:46.934059Z",
"url": "https://files.pythonhosted.org/packages/4a/18/3635f19ebe13e353fd0688884b7affacbcfd2e6fcfc4d66d773314d17ef1/py_isaac-0.1.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "5adc8935d53fec74cc2497463f6547acb84b57db1c6d5bd06c70afb9fcc4d158",
"md5": "9091827e8561ecc848e8db1d23813bf9",
"sha256": "4d3bb137e5f9cbd7a811755f9dd457e111e3281a6e184470aac25f8620f2783f"
},
"downloads": -1,
"filename": "py-isaac-0.1.3.tar.gz",
"has_sig": false,
"md5_digest": "9091827e8561ecc848e8db1d23813bf9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 27494,
"upload_time": "2025-02-15T01:09:48",
"upload_time_iso_8601": "2025-02-15T01:09:48.156648Z",
"url": "https://files.pythonhosted.org/packages/5a/dc/8935d53fec74cc2497463f6547acb84b57db1c6d5bd06c70afb9fcc4d158/py-isaac-0.1.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-15 01:09:48",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "n1teshy",
"github_project": "py-isaac",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "yapper-tts",
"specs": []
},
{
"name": "pyreadline3",
"specs": []
},
{
"name": "py-listener",
"specs": []
},
{
"name": "psutil",
"specs": []
},
{
"name": "rich",
"specs": []
}
],
"lcname": "py-isaac"
}
JSON
