academia-mcp 1.10.5__tar.gz → 1.10.7__tar.gz

{academia_mcp-1.10.5 → academia_mcp-1.10.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.10.5
+Version: 1.10.7
 Summary: MCP server that provides different tools to search for scientific publications
 Author-email: Ilya Gusev <phoenixilya@gmail.com>
 Project-URL: Homepage, https://github.com/IlyaGusev/academia_mcp

{academia_mcp-1.10.5 → academia_mcp-1.10.7}/academia_mcp/tools/web_search.py

@@ -1,5 +1,5 @@
 import json
-from typing import Optional
+from typing import Optional, List, Tuple
 
 from academia_mcp.utils import post_with_retries, get_with_retries
 from academia_mcp.settings import settings
@@ -12,10 +12,29 @@ BRAVE_SEARCH_URL = "https://api.search.brave.com/res/v1/web/search"
 EXCLUDE_DOMAINS = ["chatpaper.com"]
 
 
+def _parse_domains(query: str) -> Tuple[str, List[str]]:
+    site_term = "site:"
+    if site_term not in query:
+        return query, []
+    parts = query.split()
+    query_parts = []
+    include_domains = []
+    for part in parts:
+        if not part.startswith(site_term):
+            query_parts.append(part)
+            continue
+        domain = part[len(site_term) :]
+        if domain:
+            include_domains.append(domain)
+    query = " ".join(query_parts)
+    return query, include_domains
+
+
 def web_search(
     query: str,
     limit: Optional[int] = 20,
     provider: Optional[str] = "tavily",
+    include_domains: Optional[List[str]] = None,
 ) -> str:
     """
     Search the web using Exa Search, Brave Search or Tavily and return normalized results.
@@ -29,9 +48,22 @@ def web_search(
         query: The search query, required.
         limit: The maximum number of items to return. 20 by default, maximum 25.
         provider: The provider to use. "exa", "tavily" or "brave". "tavily" by default.
+        include_domains: Optional list of domains to include in the search. None by default.
     """
     providers = ("tavily", "brave", "exa")
     assert provider in providers, "Error: provider must be either 'exa', 'tavily' or 'brave'"
+    if include_domains:
+        assert len(include_domains) > 0, "Error: include_domains should be a non-empty list"
+        assert all(
+            isinstance(domain, str) for domain in include_domains
+        ), "Error: include_domains should be a list of strings"
+
+    query, query_include_domains = _parse_domains(query)
+    if query_include_domains:
+        if include_domains:
+            include_domains.extend(query_include_domains)
+        else:
+            include_domains = query_include_domains
 
     is_tavily_available = bool(settings.TAVILY_API_KEY)
     is_exa_available = bool(settings.EXA_API_KEY)
@@ -51,16 +83,18 @@ def web_search(
 
     result = {}
     if provider == "exa":
-        result = json.loads(exa_web_search(query, limit))
+        result = json.loads(exa_web_search(query, limit, include_domains=include_domains))
     elif provider == "brave":
         result = json.loads(brave_web_search(query, limit))
     elif provider == "tavily":
-        result = json.loads(tavily_web_search(query, limit))
+        result = json.loads(tavily_web_search(query, limit, include_domains=include_domains))
     result["search_provider"] = provider
     return sanitize_output(json.dumps(result, ensure_ascii=False))
 
 
-def tavily_web_search(query: str, limit: Optional[int] = 20) -> str:
+def tavily_web_search(
+    query: str, limit: Optional[int] = 20, include_domains: Optional[List[str]] = None
+) -> str:
     """
     Search the web using Tavily and return results.
 
@@ -71,11 +105,17 @@ def tavily_web_search(query: str, limit: Optional[int] = 20) -> str:
     Args:
         query: The search query, required.
         limit: The maximum number of items to return. 20 by default, maximum 25.
+        include_domains: Optional list of domains to include in the search. None by default.
     """
     assert isinstance(query, str), "Error: Your search query must be a string"
     assert query.strip(), "Error: Your query should not be empty"
     assert isinstance(limit, int), "Error: limit should be an integer"
     assert 0 < limit <= 25, "Error: limit should be between 1 and 25"
+    if include_domains:
+        assert len(include_domains) > 0, "Error: include_domains should be a non-empty list"
+        assert all(
+            isinstance(domain, str) for domain in include_domains
+        ), "Error: include_domains should be a list of strings"
 
     key = settings.TAVILY_API_KEY or ""
     assert key, "Error: TAVILY_API_KEY is not set and no api_key was provided"
@@ -85,6 +125,8 @@ def tavily_web_search(query: str, limit: Optional[int] = 20) -> str:
         "auto_parameters": True,
         "exclude_domains": EXCLUDE_DOMAINS,
     }
+    if include_domains:
+        payload["include_domains"] = include_domains
     response = post_with_retries(TAVILY_SEARCH_URL, payload, key)
     results = response.json()["results"]
     for result in results:
@@ -96,7 +138,9 @@ def tavily_web_search(query: str, limit: Optional[int] = 20) -> str:
     return sanitize_output(json.dumps({"results": results}, ensure_ascii=False))
 
 
-def exa_web_search(query: str, limit: Optional[int] = 20) -> str:
+def exa_web_search(
+    query: str, limit: Optional[int] = 20, include_domains: Optional[List[str]] = None
+) -> str:
     """
     Search the web using Exa and return results.
 
@@ -107,11 +151,17 @@ def exa_web_search(query: str, limit: Optional[int] = 20) -> str:
     Args:
         query: The search query, required.
         limit: The maximum number of items to return. 20 by default, maximum 25.
+        include_domains: Optional list of domains to include in the search. None by default.
     """
     assert isinstance(query, str), "Error: Your search query must be a string"
     assert query.strip(), "Error: Your query should not be empty"
     assert isinstance(limit, int), "Error: limit should be an integer"
     assert 0 < limit <= 25, "Error: limit should be between 1 and 25"
+    if include_domains:
+        assert len(include_domains) > 0, "Error: include_domains should be a non-empty list"
+        assert all(
+            isinstance(domain, str) for domain in include_domains
+        ), "Error: include_domains should be a list of strings"
 
     key = settings.EXA_API_KEY or ""
     assert key, "Error: EXA_API_KEY is not set and no api_key was provided"
@@ -129,6 +179,8 @@ def exa_web_search(query: str, limit: Optional[int] = 20) -> str:
             "context": False,
         },
     }
+    if include_domains:
+        payload["includeDomains"] = include_domains
 
     response = post_with_retries(EXA_SEARCH_URL, payload, key)
     results = response.json()["results"]

{academia_mcp-1.10.5 → academia_mcp-1.10.7}/academia_mcp.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.10.5
+Version: 1.10.7
 Summary: MCP server that provides different tools to search for scientific publications
 Author-email: Ilya Gusev <phoenixilya@gmail.com>
 Project-URL: Homepage, https://github.com/IlyaGusev/academia_mcp

{academia_mcp-1.10.5 → academia_mcp-1.10.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "academia-mcp"
-version = "1.10.5"
+version = "1.10.7"
 description = "MCP server that provides different tools to search for scientific publications"
 readme = "README.md"
 authors = [

{academia_mcp-1.10.5 → academia_mcp-1.10.7}/tests/test_web_search.py

@@ -30,3 +30,26 @@ def test_web_search_bug() -> None:
     )
     assert results
     assert len(results.splitlines()) == 1
+
+
+def test_web_search_include_domains() -> None:
+    results = web_search(
+        "autoregressive models path-star graphs",
+        include_domains=["wikipedia.org"],
+    )
+    assert results
+    results = json.loads(results)
+    assert results
+    assert len(results["results"]) > 0
+    assert all("wikipedia.org" in result["url"] for result in results["results"])
+
+
+def test_web_search_include_query_domains() -> None:
+    results = web_search(
+        "site:wikipedia.org autoregressive models path-star graphs",
+    )
+    assert results
+    results = json.loads(results)
+    assert results
+    assert len(results["results"]) > 0
+    assert all("wikipedia.org" in result["url"] for result in results["results"])