# kkg2001/ai-testcase-generator architecture

> Architecture diagram of the GitHub repository kkg2001/ai-testcase-generator, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/kkg2001/ai-testcase-generator
- Repository: https://github.com/kkg2001/ai-testcase-generator

## Overview

This is a Streamlit AI test-case generator. Users paste requirements or upload a PDF, DOCX, text, or Markdown document; the app can optionally retrieve QA guidance, then generates a structured suite that users can review and export. The source excerpts show the UI orchestration, document extraction, generation, provider setup, schema, and optional Chroma-backed RAG. Provider implementation supports Groq and Gemini; although the UI mentions Ollama, its implementation is not shown here.

Main entry points: [`app.py`](https://github.com/kkg2001/ai-testcase-generator/blob/main/app.py) (Streamlit workflow), [`src/generator.py`](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/generator.py) (Suite generation), [`src/llm_provider.py`](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/llm_provider.py) (LLM provider setup), [`src/rag.py`](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/rag.py) (RAG retrieval)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interface["User interface"]
  node_app["Streamlit workflow<br/>[app.py]"]
  node_results["Review and export<br/>[app.py]"]
end

subgraph group_inputs["Requirements input"]
  node_loader["Document extraction<br/>[document_loader.py]"]
end

subgraph group_generation["Test generation"]
  node_generator["Suite generation<br/>[generator.py]"]
  node_provider["LLM provider setup<br/>[llm_provider.py]"]
  node_schema["Test suite schema<br/>[models.py]"]
end

subgraph group_knowledge["Knowledge retrieval"]
  node_rag["RAG retrieval<br/>[rag.py]"]
  node_chromadb[("Chroma knowledge store<br/>[rag.py]")]
end

node_user(("QA user"))
node_groq["Groq API"]
node_gemini["Gemini API"]

node_user -->|"provides requirements"| node_app
node_app -->|"loads upload"| node_loader
node_app -->|"requests suite"| node_generator
node_app -.->|"queries context"| node_rag
node_rag -->|"queries and stores"| node_chromadb
node_generator -->|"creates model"| node_provider
node_generator -->|"validates suite"| node_schema
node_provider -->|"creates client"| node_groq
node_provider -->|"creates client"| node_gemini
node_generator -->|"returns suite"| node_app
node_app -->|"renders and exports"| node_results
node_generator -->|"converts to dataframe"| node_results
node_results -->|"presents suite"| node_user

click node_app "https://github.com/kkg2001/ai-testcase-generator/blob/main/app.py"
click node_results "https://github.com/kkg2001/ai-testcase-generator/blob/main/app.py"
click node_loader "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/document_loader.py"
click node_generator "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/generator.py"
click node_provider "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/llm_provider.py"
click node_schema "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/models.py"
click node_rag "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/rag.py"
click node_chromadb "https://github.com/kkg2001/ai-testcase-generator/blob/main/src/rag.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_app,node_results,node_user toneBlue
class node_loader toneAmber
class node_generator,node_provider,node_schema,node_groq,node_gemini toneMint
class node_rag,node_chromadb toneRose
```

## Components

### User interface

- [Streamlit workflow](https://github.com/kkg2001/ai-testcase-generator/blob/main/app.py): `app.py`
- [Review and export](https://github.com/kkg2001/ai-testcase-generator/blob/main/app.py): `app.py`

### Requirements input

- [Document extraction](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/document_loader.py): `src/document_loader.py`

### Test generation

- [Suite generation](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/generator.py): `src/generator.py`
- [LLM provider setup](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/llm_provider.py): `src/llm_provider.py`
- [Test suite schema](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/models.py): `src/models.py`

### Knowledge retrieval

- [RAG retrieval](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/rag.py): `src/rag.py`
- [Chroma knowledge store](https://github.com/kkg2001/ai-testcase-generator/blob/main/src/rag.py): `src/rag.py`

### Other components

- **QA user**
- **Groq API**
- **Gemini API**

## Connections

- QA user → Streamlit workflow: provides requirements (evidence: `app.py`)
- Streamlit workflow → Document extraction: loads upload (evidence: `app.py`)
- Streamlit workflow → Suite generation: requests suite (evidence: `app.py`)
- Streamlit workflow → RAG retrieval: queries context (evidence: `app.py`)
- RAG retrieval → Chroma knowledge store: queries and stores (evidence: `src/rag.py`)
- Suite generation → LLM provider setup: creates model (evidence: `src/generator.py`)
- Suite generation → Test suite schema: validates suite (evidence: `src/generator.py`)
- LLM provider setup → Groq API: creates client (evidence: `src/llm_provider.py`)
- LLM provider setup → Gemini API: creates client (evidence: `src/llm_provider.py`)
- Suite generation → Streamlit workflow: returns suite (evidence: `app.py`)
- Streamlit workflow → Review and export: renders and exports (evidence: `app.py`)
- Suite generation → Review and export: converts to dataframe (evidence: `app.py`)
- Review and export → QA user: presents suite (evidence: `app.py`)
