# karansharma18/langchain-projects architecture

> Architecture diagram of the GitHub repository karansharma18/langchain-projects, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-07). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/karansharma18/langchain-projects
- Repository: https://github.com/karansharma18/langchain-projects

## Overview

This repository is a small set of LangChain examples, not one integrated application. Users can run a prompt-to-model chain, a structured web-search agent, or an e-commerce tool-calling agent. The commerce examples look duplicated: both define a local product-price lookup and tier-discount tool, then let Gemini choose and invoke tools in a bounded loop. The README gives no architectural detail; edges below rely on the shown source excerpts.

Main entry points: [`basic_langchain_agent.py`](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py) (Search runner), [`main.py`](https://github.com/karansharma18/langchain-projects/blob/main/main.py) (Commerce runner), [`agent_loop_langchain_tool_calling.py`](https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py) (Loop runner), [`basic_langchain_prompt.py`](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py) (Prompt runner)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_prompt["Prompt example"]
  node_prompt_entry["Prompt runner"]
  node_prompt_template["Chat template"]
  node_prompt_model["Prompt model"]
end

subgraph group_research["Search agent"]
  node_research_entry["Search runner"]
  node_research_agent["Structured agent"]
  node_response_schema["Response schema"]
end

subgraph group_commerce["Commerce agent"]
  node_commerce_entry["Commerce runner<br/>[main.py]"]
  node_price_tool["Price lookup<br/>[main.py]"]
  node_discount_tool["Tier discount<br/>[main.py]"]
  node_loop_entry["Loop runner"]
  node_loop_price["Price lookup"]
  node_loop_discount["Tier discount"]
end

node_user(("User"))
node_gemini["Google Gemini"]
node_tavily["Tavily search"]

node_user -->|"runs"| node_prompt_entry
node_prompt_entry -->|"creates"| node_prompt_template
node_prompt_entry -->|"composes"| node_prompt_model
node_prompt_model -->|"invokes"| node_gemini
node_user -->|"runs"| node_research_entry
node_research_entry -->|"creates and invokes"| node_research_agent
node_research_entry -->|"configures"| node_gemini
node_research_entry -->|"registers as tool"| node_tavily
node_research_entry -->|"sets format"| node_response_schema
node_user -->|"runs"| node_commerce_entry
node_commerce_entry -->|"configures and invokes"| node_gemini
node_commerce_entry -->|"dispatches"| node_price_tool
node_commerce_entry -->|"dispatches"| node_discount_tool
node_user -->|"runs"| node_loop_entry
node_loop_entry -->|"configures and invokes"| node_gemini
node_loop_entry -->|"dispatches"| node_loop_price
node_loop_entry -->|"dispatches"| node_loop_discount

click node_prompt_entry "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py"
click node_prompt_template "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py"
click node_prompt_model "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py"
click node_research_entry "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py"
click node_research_agent "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py"
click node_response_schema "https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py"
click node_commerce_entry "https://github.com/karansharma18/langchain-projects/blob/main/main.py"
click node_price_tool "https://github.com/karansharma18/langchain-projects/blob/main/main.py"
click node_discount_tool "https://github.com/karansharma18/langchain-projects/blob/main/main.py"
click node_loop_entry "https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py"
click node_loop_price "https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py"
click node_loop_discount "https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.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_prompt_entry,node_prompt_template,node_prompt_model,node_user toneBlue
class node_research_entry,node_research_agent,node_response_schema toneAmber
class node_commerce_entry,node_price_tool,node_discount_tool,node_loop_entry,node_loop_price,node_loop_discount toneMint
class node_gemini,node_tavily toneIndigo
```

## Components

### Prompt example

- [Prompt runner](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py): `basic_langchain_prompt.py`
- [Chat template](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py): `basic_langchain_prompt.py`
- [Prompt model](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_prompt.py): `basic_langchain_prompt.py`

### Search agent

- [Search runner](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py): `basic_langchain_agent.py`
- [Structured agent](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py): `basic_langchain_agent.py`
- [Response schema](https://github.com/karansharma18/langchain-projects/blob/main/basic_langchain_agent.py): `basic_langchain_agent.py`

### Commerce agent

- [Commerce runner](https://github.com/karansharma18/langchain-projects/blob/main/main.py): `main.py`
- [Price lookup](https://github.com/karansharma18/langchain-projects/blob/main/main.py): `main.py`
- [Tier discount](https://github.com/karansharma18/langchain-projects/blob/main/main.py): `main.py`
- [Loop runner](https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py): `agent_loop_langchain_tool_calling.py`
- [Price lookup](https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py): `agent_loop_langchain_tool_calling.py`
- [Tier discount](https://github.com/karansharma18/langchain-projects/blob/main/agent_loop_langchain_tool_calling.py): `agent_loop_langchain_tool_calling.py`

### Other components

- **User**
- **Google Gemini**
- **Tavily search**

## Connections

- User → Prompt runner: runs (evidence: `basic_langchain_prompt.py`)
- Prompt runner → Chat template: creates (evidence: `basic_langchain_prompt.py`)
- Prompt runner → Prompt model: composes (evidence: `basic_langchain_prompt.py`)
- Prompt model → Google Gemini: invokes (evidence: `basic_langchain_prompt.py`)
- User → Search runner: runs (evidence: `basic_langchain_agent.py`)
- Search runner → Structured agent: creates and invokes (evidence: `basic_langchain_agent.py`)
- Search runner → Google Gemini: configures (evidence: `basic_langchain_agent.py`)
- Search runner → Tavily search: registers as tool (evidence: `basic_langchain_agent.py`)
- Search runner → Response schema: sets format (evidence: `basic_langchain_agent.py`)
- User → Commerce runner: runs (evidence: `main.py`)
- Commerce runner → Google Gemini: configures and invokes (evidence: `main.py`)
- Commerce runner → Price lookup: dispatches (evidence: `main.py`)
- Commerce runner → Tier discount: dispatches (evidence: `main.py`)
- User → Loop runner: runs (evidence: `agent_loop_langchain_tool_calling.py`)
- Loop runner → Google Gemini: configures and invokes (evidence: `agent_loop_langchain_tool_calling.py`)
- Loop runner → Price lookup: dispatches (evidence: `agent_loop_langchain_tool_calling.py`)
- Loop runner → Tier discount: dispatches (evidence: `agent_loop_langchain_tool_calling.py`)
