# pingcap-kb-source

Knowledge graph for PingCAP codebase, enabling on-demand and progressive knowledge retrieval for agents.

## Purpose

- Build and maintain knowledge graph for PingCAP repositories
- Provide semantic retrieval interface for AI agents

## Organizations

**pingcap**, **tidbcloud**, **tikv**

## Architecture

```
knowledge/
├── meta/                    # Global metadata files
│   └── repos-*.json         # Repository lists
├── deployment/              # Deployment knowledge artifacts
│   ├── deployment-*.json    # Deployment graphs and repo data
│   └── *.md                 # Deployment analyses and catalogs
└── repos/                   # Detailed repo info (by repo name)
```

## Meta Files

### Repository Lists

| File | Description | Count |
|------|-------------|-------|
| `repos-public.json` | Public repositories | 207 |
| `repos-private.json` | Private repositories | 300+ |
| `repos-forks.json` | Forked repositories | 84 |

### Deployment Related

| File | Description |
|------|-------------|
| `deployment/deployment-repos.json` | Core deployment definition repos (5) |
| `deployment/deployment-repos.md` | Deployment repo documentation |
| `deployment/deployment-repos-non-tidbcloud.json` | Non-TiDB Cloud deployment repos (2) |

#### Deployment Repos

1. **aws-shared-cd** - EKS/ACK Free Tier deployment (Pulumi)
2. **app-delivery** - TiDB Cloud application service delivery
3. **auto-deploy** - Auto deployment
4. **infra-cd** - Infra API + Multi-Account management (Pulumi)
5. **oneconfig** - Configuration definition repository

## Usage

```bash
# Find specific repo
jq '.[] | select(.name == "aws-shared-cd")' knowledge/meta/repos-private.json

# List all deployment repos
jq '.[].name' knowledge/deployment/deployment-repos.json

# Count repos by language
jq -r '.[].language' knowledge/meta/repos-private.json | sort | uniq -c | sort -rn
```
