[PR #600] [CLOSED] feat: Add Free Web Search MCP Agent — zero-cost Search-First LLM tutorial #12388

New Issue

GiteaMirror · 2026-06-20T03:04:08-05:00

GiteaMirror commented

2026-06-20 03:04:08 -05:00

📋 Pull Request Information

Original PR: https://github.com/Shubhamsaboo/awesome-llm-apps/pull/600
Author: @wd041216-bit
Created: 3/18/2026
Status: ❌ Closed

Base: main ← Head: add-free-web-search-ultimate

📝 Commits (3)

052d78b Add free-web-search-ultimate: zero-cost Search-First MCP server
986bfd9 feat: add complete free_web_search_mcp_agent tutorial code
b40d075 docs: add Glama AAA badge to free-web-search-ultimate agent README

📊 Changes

4 files changed (+161 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
➕ mcp_ai_agents/free_web_search_mcp_agent/README.md (+59 -0)
➕ mcp_ai_agents/free_web_search_mcp_agent/requirements.txt (+5 -0)
➕ mcp_ai_agents/free_web_search_mcp_agent/search_agent.py (+96 -0)

📄 Description

🔍 Free Web Search MCP Agent — Search-First LLM Tutorial

This PR adds a complete, self-contained tutorial demonstrating how to build a Search-First LLM Agent using the Model Context Protocol (MCP) and DuckDuckGo free search.

📁 What's included: `mcp_ai_agents/free_web_search_mcp_agent/`

File	Description
`search_agent.py`	Complete Streamlit app with Search-First LLM flow
`requirements.txt`	All dependencies (agno, duckduckgo-search, streamlit)
`README.md`	Setup guide, architecture diagram, usage examples

🚀 How it works

# The agent ALWAYS searches before answering
search_results = duckduckgo_search(query)
webpage_content = extract_content(search_results[0].url)
answer = llm.reason(context=webpage_content, question=query)

✨ Key Features

Zero-Cost Search: DuckDuckGo — no API key needed for search
Search-First Paradigm: LLM is instructed to always retrieve real-time data first
Eliminates Hallucinations: Grounds every answer in current web content
MCP Integration: Uses the standard Model Context Protocol
Easy Setup: pip install -r requirements.txt && streamlit run search_agent.py

🧪 Run it yourself

git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
cd awesome-llm-apps/mcp_ai_agents/free_web_search_mcp_agent
pip install -r requirements.txt
export OPENAI_API_KEY=your_key_here
streamlit run search_agent.py

Powered by: free-web-search-ultimate — zero-cost MCP server for real-time web search

This addresses the feedback from the previous PR #597 — the tutorial now includes all runnable code in its own folder.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/Shubhamsaboo/awesome-llm-apps/pull/600 **Author:** [@wd041216-bit](https://github.com/wd041216-bit) **Created:** 3/18/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `add-free-web-search-ultimate` --- ### 📝 Commits (3) - [`052d78b`](https://github.com/Shubhamsaboo/awesome-llm-apps/commit/052d78bc5417251a7ca954ca70b366160691150b) Add free-web-search-ultimate: zero-cost Search-First MCP server - [`986bfd9`](https://github.com/Shubhamsaboo/awesome-llm-apps/commit/986bfd9b6888f3c55ee792718f633fd8414070bb) feat: add complete free_web_search_mcp_agent tutorial code - [`b40d075`](https://github.com/Shubhamsaboo/awesome-llm-apps/commit/b40d075545f2134036cbc2682648a4d384123c5b) docs: add Glama AAA badge to free-web-search-ultimate agent README ### 📊 Changes **4 files changed** (+161 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) ➕ `mcp_ai_agents/free_web_search_mcp_agent/README.md` (+59 -0) ➕ `mcp_ai_agents/free_web_search_mcp_agent/requirements.txt` (+5 -0) ➕ `mcp_ai_agents/free_web_search_mcp_agent/search_agent.py` (+96 -0) </details> ### 📄 Description ## 🔍 Free Web Search MCP Agent — Search-First LLM Tutorial This PR adds a **complete, self-contained tutorial** demonstrating how to build a **Search-First LLM Agent** using the Model Context Protocol (MCP) and DuckDuckGo free search. ### 📁 What's included: `mcp_ai_agents/free_web_search_mcp_agent/` | File | Description | |------|-------------| | `search_agent.py` | Complete Streamlit app with Search-First LLM flow | | `requirements.txt` | All dependencies (agno, duckduckgo-search, streamlit) | | `README.md` | Setup guide, architecture diagram, usage examples | ### 🚀 How it works ```python # The agent ALWAYS searches before answering search_results = duckduckgo_search(query) webpage_content = extract_content(search_results[0].url) answer = llm.reason(context=webpage_content, question=query) ``` ### ✨ Key Features - **Zero-Cost Search**: DuckDuckGo — no API key needed for search - **Search-First Paradigm**: LLM is instructed to always retrieve real-time data first - **Eliminates Hallucinations**: Grounds every answer in current web content - **MCP Integration**: Uses the standard Model Context Protocol - **Easy Setup**: `pip install -r requirements.txt && streamlit run search_agent.py` ### 🧪 Run it yourself ```bash git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git cd awesome-llm-apps/mcp_ai_agents/free_web_search_mcp_agent pip install -r requirements.txt export OPENAI_API_KEY=your_key_here streamlit run search_agent.py ``` **Powered by**: [free-web-search-ultimate](https://github.com/wd041216-bit/free-web-search-ultimate) — zero-cost MCP server for real-time web search --- *This addresses the feedback from the previous PR #597 — the tutorial now includes all runnable code in its own folder.* --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

GiteaMirror added the pull-request label 2026-06-20 03:04:08 -05:00

GiteaMirror closed this issue

2026-06-20 03:04:09 -05:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/awesome-llm-apps#12388

[PR #600] [CLOSED] feat: Add Free Web Search MCP Agent — zero-cost Search-First LLM tutorial #12388