# professional_rag

**Repository Path**: wpcaishaoyu/professional_rag

## Basic Information

- **Project Name**: professional_rag
- **Description**: llm-wiki功能项目
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: feature/20260709/rag-knowledge-pipeline
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-07-10
- **Last Updated**: 2026-07-10

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Professional RAG

A wiki-first professional knowledge base built with Java 17, Gradle, and Spring Boot.

The project follows a simple memory model:

- Preserve original or lightly transformed material in `raw/`.
- Distill durable knowledge into linked Markdown pages under `wiki/`.
- Keep source references in wiki frontmatter.
- Use the CLI to initialize, ingest, query, and lint the repository.

## Commands

```bash
./gradlew bootRun --args="init"
./gradlew bootRun --args="ingest path/to/note.md"
./gradlew bootRun --args="query agent memory"
./gradlew bootRun --args="lint"
```

Use `--root` to manage a different knowledge-base directory:

```bash
./gradlew bootRun --args="--root /path/to/kb lint"
```

## Workflow

1. Run `init` to create or repair the scaffold.
2. Run `ingest` for local Markdown or text sources.
3. Edit the generated wiki draft into a source-grounded page.
4. Link related pages.
5. Run `lint` before finishing a curation pass.

## Development

```bash
./gradlew test
```