oricore
mcp
Uyari
Health Uyari
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 7 GitHub stars
Code Gecti
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
Purpose
This package is an AI engine library that allows developers to quickly integrate various AI models into their applications, featuring built-in tools for file operations, shell commands, and network requests.
Security Assessment
The tool poses a Medium security risk. While the automated code scan did not find dangerous patterns or hardcoded secrets, the library's own documentation explicitly mentions having built-in capabilities for shell command execution, file read/write operations, and network requests. This provides a broad attack surface if the tool is misconfigured or if untrusted inputs are passed to the AI engine. Additionally, the project currently has very low community visibility, meaning the codebase has not been widely scrutinized by independent security researchers.
Quality Assessment
The project is actively maintained, with its most recent code push happening today. It utilizes the permissive and standard MIT license. However, community trust and visibility are currently very low, as indicated by having only 7 GitHub stars. This suggests it is likely an early-stage or niche project rather than a battle-tested community standard.
Verdict
Use with caution: the library is active and licensed properly, but its low community adoption and highly privileged built-in tool execution capabilities warrant careful security review before deploying in production.
This package is an AI engine library that allows developers to quickly integrate various AI models into their applications, featuring built-in tools for file operations, shell commands, and network requests.
Security Assessment
The tool poses a Medium security risk. While the automated code scan did not find dangerous patterns or hardcoded secrets, the library's own documentation explicitly mentions having built-in capabilities for shell command execution, file read/write operations, and network requests. This provides a broad attack surface if the tool is misconfigured or if untrusted inputs are passed to the AI engine. Additionally, the project currently has very low community visibility, meaning the codebase has not been widely scrutinized by independent security researchers.
Quality Assessment
The project is actively maintained, with its most recent code push happening today. It utilizes the permissive and standard MIT license. However, community trust and visibility are currently very low, as indicated by having only 7 GitHub stars. This suggests it is likely an early-stage or niche project rather than a battle-tested community standard.
Verdict
Use with caution: the library is active and licensed properly, but its low community adoption and highly privileged built-in tool execution capabilities warrant careful security review before deploying in production.
Powerful AI Engine Library - Build Smart Assistants in 5 Lines of Code
README.md
What is OriCore?
OriCore is a fully-featured AI Engine Library that lets you easily integrate powerful AI capabilities into any application.
- Support for 40+ AI providers (OpenAI, Claude, DeepSeek, Zhipu AI, etc.)
- Complete built-in tool system (file read/write, code search, shell commands, network requests)
- Session management + Context compression
- MCP protocol + Skill system for unlimited extensions
- 5 professional interaction modes (brainstorm, plan, review, debug, default)
Build AI-powered products with just 5 lines of code:
import { createEngine } from 'oricore';
const engine = createEngine({ productName: 'MyApp', version: '1.0.0' });
await engine.initialize({
model: 'deepseek/deepseek-chat',
provider: {
deepseek: {
options: { apiKey: 'your-key' }
}
}
});
const result = await engine.sendMessage({ message: 'Analyze this project structure', write: true });
console.log(result.data.text);
Quick Start
npm install oricore ai
import { createEngine } from 'oricore';
const engine = createEngine({ productName: 'MyAIAssistant', version: '1.0.0' });
await engine.initialize({
model: 'deepseek/deepseek-chat',
provider: {
deepseek: {
options: { apiKey: 'your-api-key' }
}
},
});
const result = await engine.sendMessage({
message: 'Create a TypeScript function to calculate Fibonacci',
write: true,
});
console.log(result.data.text);
await engine.shutdown();
Why OriCore?
| What You Need | OriCore | Others |
|---|---|---|
| Build an AI assistant fast | ✅ 5 lines of code | ❌ 100+ lines of setup |
| Work with any AI provider | ✅ 40+ providers unified | ⚠️ Provider-specific SDKs |
| Ready-to-use tools | ✅ File, shell, search, web | ❌ Build everything yourself |
| Save on API costs | ✅ Auto context compression | ❌ Pay for full context every time |
| Extend with plugins | ✅ MCP + Skills | ❌ Complex or impossible |
| Switch AI providers | ✅ Change 1 line of code | ❌ Rewrite integration code |
Built for developers who need to ship AI features fast:
- 5-minute setup - Get a working AI assistant with file access, code search, and shell commands
- No vendor lock-in - Switch between OpenAI, Claude, DeepSeek, etc. anytime
- Production-ready - Built-in session persistence, cost tracking, and error handling
- Extensible - Add custom tools via MCP servers or local skill files
- AI-aware - Context compression and specialized modes for different tasks
Built-in Tools
| Tool | Function |
|---|---|
read |
Read files (text, images, PDF) |
write |
Write files |
edit |
Edit files (search & replace) |
glob |
Find files by pattern |
grep |
Search file contents |
bash |
Execute shell commands |
fetch |
Make HTTP requests |
task |
Launch specialized agents |
todo |
Track task progress |
askUserQuestion |
Interactive Q&A with users |
Use Cases
- Build Custom AI Assistants - Chatbots, customer service systems
- IDE Integration - Code assistants, smart autocomplete
- Code Review - Automated code review systems
- Debugging Tools - Intelligent error diagnosis
- Education Platforms - Programming teaching assistants
- Documentation Generation - Automated documentation generation
Documentation
Getting Started
- 5-Minute Quick Start - Get up to speed with OriCore
- Tutorials - Practical examples
Core Features
- API Reference - Complete API documentation
- Configuration Guide - All configuration options
- Tools System - Built-in tools guide
- Interaction Modes - 5 professional modes
Advanced Features
- Session Management - Persistence & context compression
- Event System - Message bus & events
- MCP Integration - MCP protocol support
- Skill System - Custom skill loading
- Approval System - Tool execution permission control
Acknowledgments
This project references the core architecture of the following excellent project:
neovate-code - Core AI engine architecture
OriCore has been refactored and streamlined on this foundation, removing UI, CLI, and other peripheral features to focus on providing a powerful and standalone AI engine library that can be easily integrated into any project.
License
MIT © lyw405
Support
- GitHub: lyw405/oricore
- Issues: Report a problem
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi