diff --git a/libs/miroflow-tools/pyproject.toml b/libs/miroflow-tools/pyproject.toml
index 4213ce39..ea281be1 100644
--- a/libs/miroflow-tools/pyproject.toml
+++ b/libs/miroflow-tools/pyproject.toml
@@ -18,7 +18,8 @@ dependencies = [
     "markitdown-mcp>=0.0.1a3",
     "google-genai",
     "aiohttp",
-    "redis"
+    "redis",
+    "tavily-python>=0.5.0"
 ]
 
 [build-system]
diff --git a/libs/miroflow-tools/src/miroflow_tools/mcp_servers/tavily_mcp_server.py b/libs/miroflow-tools/src/miroflow_tools/mcp_servers/tavily_mcp_server.py
new file mode 100644
index 00000000..14585855
--- /dev/null
+++ b/libs/miroflow-tools/src/miroflow_tools/mcp_servers/tavily_mcp_server.py
@@ -0,0 +1,120 @@
+# Copyright (c) 2025 MiroMind
+# This source code is licensed under the Apache 2.0 License.
+
+"""
+Tavily-backed MCP server providing a tavily_search tool as a parallel
+alternative to the Serper-backed google_search tool.
+"""
+
+import json
+import os
+
+import httpx
+from mcp.server.fastmcp import FastMCP
+from tavily import TavilyClient
+from tenacity import (
+    retry,
+    retry_if_exception_type,
+    stop_after_attempt,
+    wait_exponential,
+)
+
+# Obtain from https://app.tavily.com
+TAVILY_API_KEY = os.getenv("TAVILY_API_KEY", "")
+
+# Initialize FastMCP server
+mcp = FastMCP("tavily-mcp-server")
+
+
+@retry(
+    stop=stop_after_attempt(3),
+    wait=wait_exponential(multiplier=1, min=4, max=10),
+    retry=retry_if_exception_type((httpx.ConnectError, httpx.TimeoutException)),
+)
+def _tavily_search_request(client: TavilyClient, **kwargs) -> dict:
+    """Execute a Tavily search with retry logic."""
+    return client.search(**kwargs)
+
+
+@mcp.tool()
+def tavily_search(
+    query: str,
+    max_results: int = 10,
+    search_depth: str = "basic",
+    topic: str = "general",
+    include_domains: list[str] | None = None,
+    exclude_domains: list[str] | None = None,
+    time_range: str | None = None,
+):
+    """
+    Tool to perform web searches via Tavily API and retrieve rich results.
+
+    It returns organic search results with titles, URLs, content snippets,
+    and relevance scores.
+
+    Args:
+        query: Search query string (max 400 characters recommended)
+        max_results: Number of results to return (default: 10)
+        search_depth: Search depth - 'basic' (fast, 1 credit) or 'advanced' (thorough, 2 credits)
+        topic: Search topic category - 'general', 'news', or 'finance'
+        include_domains: Optional list of domains to restrict results to
+        exclude_domains: Optional list of domains to exclude from results
+        time_range: Optional time range filter ('day', 'week', 'month', 'year')
+
+    Returns:
+        JSON string containing search results and metadata.
+    """
+    # Check for API key
+    if not TAVILY_API_KEY:
+        return json.dumps(
+            {
+                "success": False,
+                "error": "TAVILY_API_KEY environment variable not set",
+                "results": [],
+            },
+            ensure_ascii=False,
+        )
+
+    # Validate required parameter
+    if not query or not query.strip():
+        return json.dumps(
+            {
+                "success": False,
+                "error": "Search query 'query' is required and cannot be empty",
+                "results": [],
+            },
+            ensure_ascii=False,
+        )
+
+    try:
+        client = TavilyClient(api_key=TAVILY_API_KEY)
+
+        # Build keyword arguments
+        kwargs = {
+            "query": query.strip(),
+            "max_results": max_results,
+            "search_depth": search_depth,
+            "topic": topic,
+        }
+
+        if include_domains:
+            kwargs["include_domains"] = include_domains
+        if exclude_domains:
+            kwargs["exclude_domains"] = exclude_domains
+        if time_range:
+            kwargs["time_range"] = time_range
+
+        # Make the API request with retry
+        data = _tavily_search_request(client, **kwargs)
+
+        return json.dumps(data, ensure_ascii=False)
+
+    except Exception as e:
+        return json.dumps(
+            {"success": False, "error": f"Unexpected error: {str(e)}", "results": []},
+            ensure_ascii=False,
+        )
+
+
+if __name__ == "__main__":
+    mcp.run()