academia-mcp 1.10.4__tar.gz → 1.10.6__tar.gz

{academia_mcp-1.10.4 → academia_mcp-1.10.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.10.4
+Version: 1.10.6
 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-1.10.4 → academia_mcp-1.10.6}/academia_mcp/tools/visit_webpage.py

@@ -1,6 +1,6 @@
 import re
 import json
-from typing import Optional
+from typing import Optional, Dict, Any, cast
 
 from markdownify import markdownify  # type: ignore
 
@@ -14,7 +14,7 @@ AVAILABLE_PROVIDERS = ("basic", "exa", "tavily")
 ERROR_MESSAGE = "Failed to get content from the page. Try to use another provider."
 
 
-def _exa_visit_webpage(url: str) -> str:
+def _exa_visit_webpage(url: str) -> Dict[str, Any]:
     key = settings.EXA_API_KEY or ""
     assert key, "Error: EXA_API_KEY is not set and no api_key was provided"
     payload = {
@@ -24,11 +24,11 @@ def _exa_visit_webpage(url: str) -> str:
     response = post_with_retries(EXA_CONTENTS_URL, payload=payload, api_key=key)
     results = response.json()["results"]
     if not results:
-        return json.dumps({"id": url, "error": ERROR_MESSAGE})
-    return sanitize_output(json.dumps(results[0]))
+        return {"error": ERROR_MESSAGE}
+    return cast(Dict[str, Any], results[0])
 
 
-def _tavily_visit_webpage(url: str) -> str:
+def _tavily_visit_webpage(url: str) -> Dict[str, Any]:
     key = settings.TAVILY_API_KEY or ""
     assert key, "Error: TAVILY_API_KEY is not set and no api_key was provided"
     payload = {
@@ -37,8 +37,25 @@ def _tavily_visit_webpage(url: str) -> str:
     response = post_with_retries(TAVILY_EXTRACT_URL, payload=payload, api_key=key)
     results = response.json()["results"]
     if not results:
-        return json.dumps({"id": url, "error": ERROR_MESSAGE})
-    return sanitize_output(json.dumps(results[0]["raw_content"]))
+        return {"error": ERROR_MESSAGE}
+    return {"text": results[0]["raw_content"]}
+
+
+def _basic_visit_webpage(url: str) -> Dict[str, Any]:
+    try:
+        response = get_with_retries(url)
+        content_type = response.headers.get("content-type", "").lower()
+        if not content_type or (
+            not content_type.startswith("text/") and "html" not in content_type
+        ):
+            if settings.EXA_API_KEY:
+                return _exa_visit_webpage(url)
+            return {"error": f"Unsupported content-type: {content_type or 'unknown'}"}
+        markdown_content = markdownify(response.text).strip()
+        markdown_content = re.sub(r"\n{3,}", "\n\n", markdown_content)
+        return {"text": markdown_content}
+    except Exception as e:
+        return {"error": str(e) + "\n" + ERROR_MESSAGE}
 
 
 def visit_webpage(url: str, provider: Optional[str] = "basic") -> str:
@@ -59,24 +76,12 @@ def visit_webpage(url: str, provider: Optional[str] = "basic") -> str:
     ), f"Invalid provider: {provider}. Available providers: {AVAILABLE_PROVIDERS}"
 
     if provider == "exa" and settings.EXA_API_KEY:
-        return _exa_visit_webpage(url)
+        result = _exa_visit_webpage(url)
     elif provider == "tavily" and settings.TAVILY_API_KEY:
-        return _tavily_visit_webpage(url)
+        result = _tavily_visit_webpage(url)
     else:
-        provider = "basic"
+        result = _basic_visit_webpage(url)
 
-    assert provider == "basic"
-    try:
-        response = get_with_retries(url)
-    except Exception as e:
-        return json.dumps({"id": url, "error": str(e)})
-    content_type = response.headers.get("content-type", "").lower()
-    if not content_type or (not content_type.startswith("text/") and "html" not in content_type):
-        if settings.EXA_API_KEY:
-            return _exa_visit_webpage(url)
-        return json.dumps(
-            {"id": url, "error": f"Unsupported content-type: {content_type or 'unknown'}"}
-        )
-    markdown_content = markdownify(response.text).strip()
-    markdown_content = re.sub(r"\n{3,}", "\n\n", markdown_content)
-    return sanitize_output(json.dumps({"id": url, "text": markdown_content}))
+    result["id"] = url
+    result["provider"] = provider
+    return sanitize_output(json.dumps(result, ensure_ascii=False))

{academia_mcp-1.10.4 → academia_mcp-1.10.6}/academia_mcp/tools/web_search.py

@@ -1,5 +1,5 @@
 import json
-from typing import Optional
+from typing import Optional, List
 
 from academia_mcp.utils import post_with_retries, get_with_retries
 from academia_mcp.settings import settings
@@ -16,6 +16,7 @@ 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 +30,15 @@ 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"
 
     is_tavily_available = bool(settings.TAVILY_API_KEY)
     is_exa_available = bool(settings.EXA_API_KEY)
@@ -51,16 +58,18 @@ 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 +80,17 @@ 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 +100,8 @@ 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 +113,9 @@ 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 +126,17 @@ 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 +154,8 @@ 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-1.10.4 → academia_mcp-1.10.6}/academia_mcp.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.10.4
+Version: 1.10.6
 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-1.10.4 → academia_mcp-1.10.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "academia-mcp"
-version = "1.10.4"
+version = "1.10.6"
 description = "MCP server that provides different tools to search for scientific publications"
 readme = "README.md"
 authors = [

{academia_mcp-1.10.4 → academia_mcp-1.10.6}/tests/test_web_search.py

@@ -30,3 +30,15 @@ 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"])