# conllect-llm

**Repository Path**: one_love_op/conllect-llm

## Basic Information

- **Project Name**: conllect-llm
- **Description**: 可配置的 LLM Agent 平台
一个低代码的 AI Agent 构建与运行平台。不需要写代码，通过界面配置模型、提示词和工具，即可创建智能 Agent 并运行调试。
- **Primary Language**: Unknown
- **License**: GPL-3.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-06-02
- **Last Updated**: 2026-06-03

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# conllect-llm

`conllect-llm` is planned as a dynamic configurable LLM Agent platform.

The project combines:

- A platform-style Agent Runtime.
- A low-code configuration console.
- A unified model gateway for multiple LLM providers.
- A tool registry for agent tool calling.
- Prompt, memory, trace, and observability modules.

## MVP Loop

```text
configure model -> configure prompt -> configure tool -> create agent -> run agent -> inspect trace
```

## Documentation

- **[用户使用指南](docs/user-guide.md) — 小白也能看懂的完整配置教程（中文）**
- **[技术开发文档](docs/dev-guide.md) — 架构、数据模型、API 参考、扩展开发（中文）**
- [Architecture](docs/architecture.md) — 架构设计（英文）
- [Agent Configuration](docs/agent-config.md) — Agent 配置设计（英文）
- [API Design](docs/api.md) — API 设计（英文）
- [Implementation Roadmap](docs/implementation-roadmap.md)
- [Project Structure](docs/project-structure.md)

## First Implementation Target

The first implementation should provide:

- FastAPI backend.
- PostgreSQL storage.
- OpenAI-compatible model gateway.
- Prompt rendering.
- HTTP tool execution.
- Configurable agent runtime.
- Agent run trace storage.
- React Web console with a Playground.