academia-mcp 1.10.6__py3-none-any.whl → 1.10.8__py3-none-any.whl

academia_mcp/server.py

@@ -42,6 +42,7 @@ from academia_mcp.tools.bitflip import (
 from academia_mcp.tools.review import review_pdf_paper, download_pdf_paper
 from academia_mcp.tools.show_image import show_image, describe_image
 from academia_mcp.tools.speech_to_text import speech_to_text
+from academia_mcp.tools.yt_transcript import yt_transcript
 
 
 def configure_uvicorn_style_logging(level: int = logging.INFO) -> None:
@@ -92,6 +93,7 @@ def run(
     server.add_tool(get_latex_templates_list)
     server.add_tool(visit_webpage)
     server.add_tool(show_image)
+    server.add_tool(yt_transcript)
 
     if settings.WORKSPACE_DIR:
         server.add_tool(compile_latex)

academia_mcp/settings.py

@@ -21,6 +21,9 @@ class Settings(BaseSettings):
     DOCUMENT_QA_DOCUMENT_MAX_LENGTH: int = 200000
     DESCRIBE_IMAGE_MODEL_NAME: str = "gpt-4.1"
 
+    WEBSHARE_PROXY_USERNAME: Optional[str] = None
+    WEBSHARE_PROXY_PASSWORD: Optional[str] = None
+
     PORT: int = 5056
     WORKSPACE_DIR: Optional[Path] = None
 

academia_mcp/tools/__init__.py

@@ -16,6 +16,7 @@ from .bitflip import extract_bitflip_info, generate_research_proposals, score_re
 from .review import review_pdf_paper, download_pdf_paper, review_pdf_paper_by_url
 from .show_image import show_image, describe_image
 from .speech_to_text import speech_to_text
+from .yt_transcript import yt_transcript
 
 __all__ = [
     "arxiv_search",
@@ -45,4 +46,5 @@ __all__ = [
     "show_image",
     "describe_image",
     "speech_to_text",
+    "yt_transcript",
 ]

academia_mcp/tools/web_search.py

@@ -1,5 +1,5 @@
 import json
-from typing import Optional, List
+from typing import Optional, List, Tuple
 
 from academia_mcp.utils import post_with_retries, get_with_retries
 from academia_mcp.settings import settings
@@ -12,6 +12,24 @@ 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,
@@ -40,6 +58,13 @@ def web_search(
             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)
     is_brave_available = bool(settings.BRAVE_API_KEY)

academia_mcp/tools/yt_transcript.py

@@ -0,0 +1,30 @@
+from youtube_transcript_api import YouTubeTranscriptApi
+from youtube_transcript_api.proxies import WebshareProxyConfig
+
+from academia_mcp.settings import settings
+
+
+def yt_transcript(video_url: str) -> str:
+    """
+    Tool to fetch the transcript of a YouTube video given its URL.
+
+    Returns a transcript of the video as a single string.
+
+    Args:
+        video_url (str): YouTube video URL.
+    """
+    if "youtu.be" in video_url:
+        video_id = video_url.strip().split("youtu.be/")[1]
+    else:
+        video_id = video_url.strip().split("v=")[-1]
+    video_id = video_id.split("?")[0]
+    proxy_config = None
+    if settings.WEBSHARE_PROXY_USERNAME and settings.WEBSHARE_PROXY_PASSWORD:
+        proxy_config = WebshareProxyConfig(
+            proxy_username=settings.WEBSHARE_PROXY_USERNAME,
+            proxy_password=settings.WEBSHARE_PROXY_PASSWORD,
+        )
+    api = YouTubeTranscriptApi(proxy_config=proxy_config)
+    transcript = api.fetch(video_id)
+    snippets = transcript.snippets
+    return "\n".join([f"{int(entry.start)}: {' '.join(entry.text.split())}" for entry in snippets])

{academia_mcp-1.10.6.dist-info → academia_mcp-1.10.8.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.10.6
+Version: 1.10.8
 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
@@ -30,6 +30,7 @@ Requires-Dist: datasets>=4.0.0
 Requires-Dist: pymupdf>=1.26.4
 Requires-Dist: pillow>=11.3.0
 Requires-Dist: pydantic-settings>=2.6.0
+Requires-Dist: youtube-transcript-api>=1.2.2
 Dynamic: license-file
 
 # Academia MCP

{academia_mcp-1.10.6.dist-info → academia_mcp-1.10.8.dist-info}/RECORD

@@ -4,12 +4,12 @@ academia_mcp/files.py,sha256=ynIt0XbU1Z7EPWkv_hVX0pGKsLlmjYv-MVJLOfi6yzs,817
 academia_mcp/llm.py,sha256=zpGkuJFf58Ofgys_fi28-47_wJ1a7sIs_yZvI1Si6z0,993
 academia_mcp/pdf.py,sha256=9PlXzHGhb6ay3ldbTdxCcTWvH4TkET3bnb64mgoh9i0,1273
 academia_mcp/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-academia_mcp/server.py,sha256=hNfQSFwgbVMNyB7ElGpJ0RCTGeRP7jb5--7KZmG8cW8,5423
-academia_mcp/settings.py,sha256=MSQYjmhZ3NDalTzu4z3ey1Aw60TrhkDcPWUTE4-iOaU,995
+academia_mcp/server.py,sha256=tZ57YkW2EcW4DRIk87n2PFZkkTjAVsVQ5lphvc1AVA4,5517
+academia_mcp/settings.py,sha256=c5s4dI8V_cWmMED-jKDmHjfdIaBcxwEK4HdHNQ3WUIg,1096
 academia_mcp/utils.py,sha256=lRlb615JJ_0d4gcFpMoBjB6w0xXcde9dFDw0LwYpSPQ,4863
 academia_mcp/latex_templates/agents4science_2025/agents4science_2025.sty,sha256=hGcEPCYBJS4vdhWvN_yEaJC4GvT_yDroI94CfY2Oguk,12268
 academia_mcp/latex_templates/agents4science_2025/agents4science_2025.tex,sha256=Tl1QkHXHRopw9VEfWrD3Layr5JP_0gIzVQjL4KXIWqc,15814
-academia_mcp/tools/__init__.py,sha256=lGUy5C4IymplHOXqOiwDD7CT4Z8aPHJqSxXo2g9qkks,1493
+academia_mcp/tools/__init__.py,sha256=Z30vULZwUeUX5nDz5wcv0znhAeBtZRa0dvz7vD8SUYE,1555
 academia_mcp/tools/anthology_search.py,sha256=rhFpJZqGLABgr0raDuH0CARBiAJNJtEI4dlMrKNHfDQ,7669
 academia_mcp/tools/arxiv_download.py,sha256=gBY0_Kz0yGtVkLMwn6GrAyfBjovZVgcSMuyy67p65Cw,10474
 academia_mcp/tools/arxiv_search.py,sha256=pzM18qrF3QL03A53w003kE7hQi3s3QKtjgw0m7K88UY,8355
@@ -23,10 +23,11 @@ academia_mcp/tools/s2.py,sha256=QX7-pbetab3Xt_1tvVPU6o5D_NAe9y6jcTGRBK1vwtY,6200
 academia_mcp/tools/show_image.py,sha256=jiJlQ53dbZ0T61OBhCT3IKVvBl9NHc6jHgWLfg5BxiE,3856
 academia_mcp/tools/speech_to_text.py,sha256=YZzMqdvunzXkpcadP_mYhm6cs4qH1Y_42SfY-7eX4O4,1601
 academia_mcp/tools/visit_webpage.py,sha256=swlFwWRzWc7-AHP2ouRZJScSTA4dHZ32fuJnA2V0lUc,3311
-academia_mcp/tools/web_search.py,sha256=2haTqnWkoISWS7VZSI0h7mW-jjWFItA8SL1FkkUxzI8,7864
-academia_mcp-1.10.6.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-academia_mcp-1.10.6.dist-info/METADATA,sha256=pwPpp2xupjkVubi5Yt6JQqknFserzyRn-0xc2MGsIYI,6311
-academia_mcp-1.10.6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-academia_mcp-1.10.6.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
-academia_mcp-1.10.6.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
-academia_mcp-1.10.6.dist-info/RECORD,,
+academia_mcp/tools/web_search.py,sha256=VphVztf2jZNT3bPJPJuTdMkKbe2-LIbSV7keKV47lac,8616
+academia_mcp/tools/yt_transcript.py,sha256=NPBVGN-LG_N6yElQYBMEnbNhSjkM3-DPeVIJwlFGqA0,1104
+academia_mcp-1.10.8.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+academia_mcp-1.10.8.dist-info/METADATA,sha256=Cw-6atEo6S0MAjBKkgsHZxw-GJCcTZWAiOSQGeftarM,6356
+academia_mcp-1.10.8.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+academia_mcp-1.10.8.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
+academia_mcp-1.10.8.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
+academia_mcp-1.10.8.dist-info/RECORD,,