The most minimalistic library for building intelligent AI Agents straight from Python Objects
Project description
Otto: The minimalistic Framework to build AI Assistants
Welcome to Otto, the most minimalistic library for building intelligent AI Assistants, adept for also handling multitenancy and secure authentication.
pip install ottoai
Example
Turn the Github SDK into an agent that can answer questions from Github live data using pandas to calculate results.
import ottoai
import openai
import pandas
from github import Github, Auth
# create an assistant
eve = ottoai.Assistant(name="eve", personality="Like Scarlett Johansson with John Oliver's wits", llm_engine=openai, model="gpt-4")
# Make your agent capable of answering questions and do all things Github, by simply passing the sdk module
# Remember. Just pass objects as skills, and Otto will figure out the rest.
eve.add_skill(Github)
eve.add_skill(pandas)
# Start a conversation and add user specific context (variable names must be interpretable, Otto will take care of the rest)
# In this case the assistant will need access token to github, this is so you can pass context dynamically (solving for multitenancy)
conversation = eve.start_conversation(user_name = "Bob", user_context_variables = {"github_api_token":"sometoken"} )
# ask a question
response = conversation.message("Who were the last 5 people to star the mindsdb/otto repo?")
As simple as that, no need to build tools or chaining, just pass objects as skills, otto figures out the rest.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file ottoai-0.0.5.tar.gz.
File metadata
- Download URL: ottoai-0.0.5.tar.gz
- Upload date:
- Size: 5.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.18
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
67983621509fc1d90b54eec32eedc68705001129fcf4798664f1e19b582e97e1
|
|
| MD5 |
f56141b05f42cc442437089674483908
|
|
| BLAKE2b-256 |
48d924e12c9b32b50b363cda2d397a7bb3e93828c43dd0ab4d29a8c865054067
|
File details
Details for the file ottoai-0.0.5-py3-none-any.whl.
File metadata
- Download URL: ottoai-0.0.5-py3-none-any.whl
- Upload date:
- Size: 6.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.18
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f6e8a511375bd5c72d1a63e8632b5590f38ea22868bb2b825396859d653e2425
|
|
| MD5 |
4255bc625ac0d3c828208a1fdb594bb7
|
|
| BLAKE2b-256 |
90ab1a6de247d02c9ee4b0f23c3f5548e00a5eff4f3906d839fa7cb6a2739555
|