AgentOS Runtime
AI Agent CLI for workplace automation — schedule meetings, generate workflows, and automate repetitive tasks using natural language.
pip install agentos-runtime
agentos init
agentos schedule "invite john@company.com to onboarding sync tomorrow at 2pm"
agentos workflow "Create a Workflow Definition for AgentOS Website to execute Enterprise Process Automation"
What it does
AgentOS Runtime gives you a terminal command (agentos) that uses an AI agent (ReAct loop with LLM function calling) to:
- Schedule Teams meetings from plain English — the agent reasons about your request, generates a rich meeting agenda, opens Outlook, and sends the invite
- Attach files (PDF, Word, invoices) to meeting invites
- Generate agentic workflows for multi-step enterprise processes (requires AgentOS backend)
Requirements
- Python 3.10+
- Windows (Outlook COM automation requires Outlook desktop installed and logged in)
- One of: Gemini API key, OpenAI API key, Anthropic API key, GCP service account, or any LiteLLM-supported provider
Install
pip install agentos-runtime
Setup
agentos init
The setup wizard asks you:
- Which LLM provider (OpenAI, Anthropic, Gemini, Vertex AI, Groq, Azure, Kimi...)
- Your API key or GCP service account path
- Your Outlook email (must be logged into Outlook desktop)
- AgentOS server URL (leave default if running locally)
Config is saved to ~/.agentos/config.json — your keys stay on your machine.
Usage
Schedule a Teams meeting
agentos schedule "invite sarah@company.com to 1:1 sync tomorrow at 3pm"
agentos schedule "onboarding kickoff for John on April 20 at 10am for 2 hours"
agentos schedule "PROCESS.AI discussion with team@co.com on 15/04 from 14:00 to 16:00" \
-a project_brief.pdf \
-a onboarding_checklist.docx
The agent will:
- Resolve the date and time
- Ask for any missing attendee emails via a popup
- Generate a full meeting agenda using the LLM
- Open Outlook and send the invite with all attachments
Generate a workflow
# Requires AgentOS backend running: python run.py
agentos workflow "onboard a new employee with HR, IT and procurement steps"
agentos workflow "laptop procurement with manager approval" --name laptop_flow
Manage config
agentos config show
agentos config set organizer_email x@co.com
agentos config set server_url https://api.yourdomain.com
commands sheet
agentos --help
Supported LLM Providers
| Provider | How to configure |
|---|---|
| OpenAI (GPT-4o, GPT-4-turbo) | API key from platform.openai.com |
| Anthropic (Claude Sonnet/Opus) | API key from console.anthropic.com |
| Google Gemini | API key from aistudio.google.com |
| Google Vertex AI | GCP service account JSON + project ID |
| Groq (LLaMA, Mixtral) | API key from console.groq.com |
| Kimi / Moonshot | API key from platform.moonshot.cn |
| Azure OpenAI | API key + endpoint URL |
Switch provider anytime: agentos config set llm_provider openai
How it works
agentos schedule "..."
│
▼
CLI loads ~/.agentos/config.json → sets env vars
│
▼
ReAct Loop (LLM reasons + calls tools):
1. get_current_date → resolves relative dates
2. ask_user_for_input → popup for missing emails
3. generate_agenda → LLM writes meeting email body
4. schedule_meeting → Outlook COM creates + sends invite
│
▼
Outlook opens live → invite delivered to attendees
License
MIT — see LICENSE
Metadata
Release files for agentos-runtime 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 | Actions |
|---|---|---|---|---|
| agentos_runtime-0.1.0.tar.gz | 17.5 kB | Without provenance | Details for agentos_runtime-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentos_runtime-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for agentos_runtime-0.1.0-py3-none-any.whl |
Total release size: 36.1 kB
Release files / agentos_runtime-0.1.0.tar.gz
| Download URL | agentos_runtime-0.1.0.tar.gz |
|---|---|
| Size | 17.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
beb65d6e9e8790b97d9ff8f948d97230f2371cb2183b7c65aeb092214df933b1
|
|
BLAKE2b-256 checksum How to use checksums |
617abfeb945f4f1ec6ade48559c1c2cdea0719d9434a42f06de1c3b04bae0932
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.11
|
Release files / agentos_runtime-0.1.0-py3-none-any.whl
| Download URL | agentos_runtime-0.1.0-py3-none-any.whl |
|---|---|
| Size | 18.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
db27d836e8e03ca14035e6047c9aad8e94b10bbcdd2cbb60831147483de4ebc4
|
|
BLAKE2b-256 checksum How to use checksums |
5f27a7054e219c81a552f60241cecfc42e0bb4e1da7dbd7102249bb8317bdae4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.11
|
Security for agentos-runtime 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.