Exorous Code

A professional AI coding agent that can execute tasks using tools and manage complex development workflows.

Prerequisites

• Python 3.10 or higher is required.

Installation

Install via PyPI:

pip install exorous

Alternatively, install directly from the source:

pip install .

Or if you are developing:

pip install -e .

Usage

Start the interactive coding agent:

exorous

On the first run, Exorous will prompt you to:

1. Select your preferred LLM provider (OpenRouter, OpenAI, Anthropic, or Gemini).
2. Enter your API key for the selected provider.

Alternatively, run a single prompt:

exorous "Refactor the main.py file"

• Multi-turn conversations with tool calling
• Configurable model settings and temperature

Built-in Tools

• File operations: read, write, edit files
• Directory operations: list directories, search with glob patterns
• Text search: grep for pattern matching
• Shell execution: run shell commands
• Web access: search and fetch web content
• Memory: store and retrieve information
• Todo: manage task lists

Context Management

• Automatic context compression when approaching token limits
• Tool output pruning to manage context size
• Token usage tracking

Safety and Approval

• Multiple approval policies: on-request, auto, never, yolo
• Dangerous command detection and blocking
• Path-based safety checks
• User confirmation prompts for mutating operations

Session Management

• Save and resume sessions
• Create checkpoints
• Persistent session storage

MCP Integration

• Connect to Model Context Protocol servers
• Use tools from MCP servers
• Support for stdio and HTTP/SSE transports

Subagents

• Specialized subagents for specific tasks
• Built-in subagents: codebase investigator, code reviewer
• Configurable subagent definitions with custom tools and limits

Loop Detection

• Detects repeating actions
• Prevents infinite loops in agent execution

Hooks System

• Execute scripts before/after agent runs
• Execute scripts before/after tool calls
• Error handling hooks
• Custom commands and scripts

Configuration

• Configurable working directory
• Tool allowlisting
• Developer and user instructions
• Shell environment policies
• MCP server configuration

Advanced Intelligence

• Semantic Search (RAG): Index the entire codebase into a local vector database (ChromaDB) for meaning-based code retrieval.
• AST-based Chunking: Intelligent code splitting for high-fidelity semantic search.
• Knowledge Graph: Full symbol map (functions, classes, imports) with cross-file relationship tracking.
• Real-time Indexing: Automatic background updates as you modify your code.
• Long-Term Memory: Persistent project-wide conventions and decisions that survive session restarts.

User Interface

• Terminal UI with formatted output
• Command interface: /help, /config, /tools, /mcp, /stats, /save, /resume, /checkpoint, /restore
• Real-time tool call visualization