AI BOSS (Brain-Operating Support System)
AI BOSS is a local AI mentor and life-assistant that runs entirely in your terminal.
It connects to any OpenAI-compatible LLM (local or remote) and helps you manage tasks, plans, diary, reminders, and personal growth through a structured workspace.
Note: This project is in active alpha development. APIs and file formats may change.
Features
- Local & Private – Works with local LLMs (Ollama, LM Studio, vLLM, llama.cpp) or any OpenAI-compatible API.
- Persistent Workspace – Keeps your profile, plans (today/tomorrow), diary, and custom files in a dedicated folder.
- AI Tools – The assistant can create/update/delete files, set timers, send system notifications, update your profile, and more.
- Preprocessors & Postprocessors – Background tasks that run before/after the main conversation (e.g., morning routine, evening reflection).
- Timer Manager – Set reminders; the AI wakes up automatically and notifies you.
- Rich Terminal UI – Beautiful markdown rendering, syntax highlighting, reasoning display, and usage/timings.
- Multilingual – Russian and English interfaces (more can be added via locale files).
- Debugging & Transparency – Optional logging of all LLM traffic, tool calls, and pre/postprocessor snapshots.
Installation
pip install ai-boss
Requires Python 3.10 or newer.
Quick Start
After installation, simply run:
ai-boss
On first launch, an onboarding wizard will guide you through:
- Choosing the interface language.
- Selecting a workspace directory (where state files are stored).
- Configuring the LLM connection (API base URL, API key, model name).
- Advanced options (logging, reasoning effort, postprocessor quiet mode, etc.).
The wizard creates config.json and a workspace/ folder with template files.
Usage
Once running, you can chat with the AI naturally. Commands start with ! (e.g., !help).
The AI can also use tools autonomously to manage your workspace.
Common Commands
| Command | Description |
|---|---|
!help |
Show all available commands |
!status |
List files managed by the AI |
!profile |
View or edit your user profile |
!system_prompt |
View or edit the system prompt |
!model |
Change the LLM model |
!language |
Switch interface language (ru/en) |
!reset |
Clear conversation history |
!redo |
Regenerate the last assistant response |
!exit |
Quit the application |
To view a file from the files/ folder directly in the terminal, type:
/files/path/to/file.md
Without a path, it lists all files.
How It Works
AI BOSS maintains a workspace with the following structure:
workspace/
├── system_prompt.txt # Core instructions for the AI
├── user_profile.txt # Information about you
├── chat_history.json # Conversation history
├── plan_today.txt # Today's plan (read-only after morning reset)
├── plan_tomorrow.txt # Tomorrow's plan (editable)
├── diary.txt # Success diary
├── timers.json # Active reminders
└── files/ # Custom files created by the AI
Before each LLM call, the system assembles context from these files and injects it as tool results.
The AI can call tools to modify the workspace.
After the main response, postprocessors run in background threads to perform tasks like updating plans or reflecting on the conversation.
Preprocessors run before the main call (e.g., morning actualization).
Configuration
Configuration is stored in config.json in the current working directory.
You can edit it manually or use interactive commands.
Key settings include:
llm: API base, key, model, context window, max tokens, reasoning effort.workspace_dir: Where state files are stored.preprocessors/postprocessors: Background task definitions.language: Interface language (ruoren).debug_traffic: Enable detailed logging tologs/.
Contributing
Contributions are welcome!
Please open an issue or pull request on GitHub.
License
This project is licensed under the Apache License 2.0 – see the LICENSE file for details.
Links
- GitHub: https://github.com/backup64back-afk/Ai-Boss
- Issues: https://github.com/backup64back-afk/Ai-Boss/issues
- Changelog: https://github.com/backup64back-afk/Ai-Boss/blob/main/CHANGELOG.md
Made with ❤️ for personal growth and productivity.
Metadata
Release files for ai-boss 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| ai_boss-0.1.0.tar.gz | 72.9 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ai_boss-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 158.1 kB
Release files / ai_boss-0.1.0.tar.gz
| Download URL | ai_boss-0.1.0.tar.gz |
|---|---|
| Size | 72.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
17eb68a1ca144dcc9956af985574d1d58d75021609b637aea5602e390cf2f9dd
|
|
BLAKE2b-256 checksum How to use checksums |
2af8c73c5063a8caaf718f956a41e8e8646f8b149323254ef3cfbd9d9906cc6b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|
Release files / ai_boss-0.1.0-py3-none-any.whl
| Download URL | ai_boss-0.1.0-py3-none-any.whl |
|---|---|
| Size | 85.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
1517a99a0d2b649942e035471f6c6aad381b82b5960e7804f26d938e84aedb7f
|
|
BLAKE2b-256 checksum How to use checksums |
85141c79a1601c9072dcd4cec9618ff478b0935f9282362789b633c4df4d456d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|
Security for ai-boss 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.