Skip to content

Latest commit

 

History

History
64 lines (44 loc) · 1.93 KB

File metadata and controls

64 lines (44 loc) · 1.93 KB

🤖 Microbots

MicroBots is a lightweight, extensible AI agent for code comprehension and controlled file edits. It integrates cleanly into automation pipelines, mounting a target directory with explicit read-only or read/write modes so LLMs can safely inspect, refactor, or generate files with least-privilege access.

🚀 Quick Start

Pre-requisites

  • Docker
  • AI LLM Provider and API Key

Install

pip install microbots

Example

from microbots import WritingBot

myWritingBot = WritingBot(
    model="azure-openai/my-gpt5",
    folder_to_mount=str("myReactApp"),
)

data = myWritingBot.run(
    "Fix the build error and make sure the build is successful.",
    timeout_in_seconds=600,
)
print(data.results)

🤖 Available Bots

Bot Description
ReadingBot Reads files and extracts information based on instructions (read-only)
WritingBot Reads and writes files based on instructions (read/write)
BrowsingBot Browses the web to gather information
LogAnalysisBot Analyzes logs for debugging
AgentBoss Orchestrates multiple bots for complex tasks

⚙️ How it works

Overall Architecture

MicroBots creates a containerized environment and mounts the specified directory, restricting permissions to read-only or read/write based on the Bot used. This ensures AI agents operate within defined boundaries, enhancing security and control over code modifications while protecting the local environment.

✨ LLM Support

Azure OpenAI Models — add environment variables in a .env file:

OPEN_AI_END_POINT=your-endpoint-url
OPEN_AI_KEY=your-api-key

📚 Links