academia-mcp 1.11.3__tar.gz → 1.11.4__tar.gz

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.11.3
+Version: 1.11.4
 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
@@ -12,16 +12,14 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: mcp>=1.10.1
 Requires-Dist: xmltodict>=0.14.0
-Requires-Dist: types-xmltodict>=0.14.0
 Requires-Dist: requests>=2.32.0
-Requires-Dist: types-requests>=2.32.0
 Requires-Dist: pypdf>=5.1.0
 Requires-Dist: beautifulsoup4>=4.12.0
-Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: markdownify==0.14.1
+Requires-Dist: types-xmltodict>=0.14.0
+Requires-Dist: types-requests>=2.32.0
+Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: acl-anthology==0.5.2
-Requires-Dist: markdown==3.7.0
-Requires-Dist: types-markdown==3.7.0.20250322
 Requires-Dist: huggingface-hub>=0.32.4
 Requires-Dist: fire>=0.7.0
 Requires-Dist: openai>=1.97.1
@@ -31,6 +29,8 @@ 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
+Requires-Dist: paddlepaddle>=3.2.0
+Requires-Dist: paddleocr>=3.2.0
 Dynamic: license-file
 
 # Academia MCP
@@ -70,12 +70,16 @@ make install
 ### Quickstart
 - Run over HTTP (default transport):
 ```bash
+python -m academia_mcp --transport streamable-http
+# OR
 uv run -m academia_mcp --transport streamable-http
 ```
 
 - Run over stdio (for local MCP clients like Claude Desktop):
 ```bash
 python -m academia_mcp --transport stdio
+# OR
+uv run -m academia_mcp --transport stdio
 ```
 
 Notes:
@@ -122,7 +126,7 @@ Availability notes:
 - Set one or more of `EXA_API_KEY`, `BRAVE_API_KEY`, `TAVILY_API_KEY` to enable `web_search` and provider tools.
 
 ### Environment variables
-Set as needed depending on which tools you use:
+Set as needed, depending on which tools you use:
 
 - `OPENROUTER_API_KEY`: required for LLM-related tools.
 - `BASE_URL`: override OpenRouter base URL.
@@ -152,7 +156,7 @@ docker run --rm -p 5056:5056 \
   academia_mcp
 ```
 
-Or use existing image: `phoenix120/academia_mcp`
+Or use existing image: [`phoenix120/academia_mcp`](https://hub.docker.com/repository/docker/phoenix120/academia_mcp)
 
 ### Examples
 - [Comprehensive report screencast (YouTube)](https://www.youtube.com/watch?v=4bweqQcN6w8)

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/README.md

@@ -35,12 +35,16 @@ make install
 ### Quickstart
 - Run over HTTP (default transport):
 ```bash
+python -m academia_mcp --transport streamable-http
+# OR
 uv run -m academia_mcp --transport streamable-http
 ```
 
 - Run over stdio (for local MCP clients like Claude Desktop):
 ```bash
 python -m academia_mcp --transport stdio
+# OR
+uv run -m academia_mcp --transport stdio
 ```
 
 Notes:
@@ -87,7 +91,7 @@ Availability notes:
 - Set one or more of `EXA_API_KEY`, `BRAVE_API_KEY`, `TAVILY_API_KEY` to enable `web_search` and provider tools.
 
 ### Environment variables
-Set as needed depending on which tools you use:
+Set as needed, depending on which tools you use:
 
 - `OPENROUTER_API_KEY`: required for LLM-related tools.
 - `BASE_URL`: override OpenRouter base URL.
@@ -117,7 +121,7 @@ docker run --rm -p 5056:5056 \
   academia_mcp
 ```
 
-Or use existing image: `phoenix120/academia_mcp`
+Or use existing image: [`phoenix120/academia_mcp`](https://hub.docker.com/repository/docker/phoenix120/academia_mcp)
 
 ### Examples
 - [Comprehensive report screencast (YouTube)](https://www.youtube.com/watch?v=4bweqQcN6w8)

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/academia_mcp/server.py

@@ -40,7 +40,7 @@ from academia_mcp.tools.bitflip import (
     score_research_proposals,
 )
 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.image_processing 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
 

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/academia_mcp/tools/__init__.py

@@ -14,7 +14,7 @@ from .web_search import web_search, tavily_web_search, exa_web_search, brave_web
 from .visit_webpage import visit_webpage
 from .bitflip import extract_bitflip_info, generate_research_proposals, score_research_proposals
 from .review import review_pdf_paper, download_pdf_paper, review_pdf_paper_by_url
-from .show_image import show_image, describe_image
+from .image_processing import show_image, describe_image
 from .speech_to_text import speech_to_text
 from .yt_transcript import yt_transcript
 

academia_mcp-1.11.3/academia_mcp/tools/show_image.py → academia_mcp-1.11.4/academia_mcp/tools/image_processing.py

@@ -1,16 +1,28 @@
+import asyncio
 import base64
-from pathlib import Path
+import contextlib
+import json
+import logging
+import os
+import threading
 from io import BytesIO
-from typing import Dict, Optional
+from pathlib import Path
 from textwrap import dedent
+from typing import Dict, List, Optional, Any
 
 import httpx
+from paddleocr import PaddleOCR  # type: ignore
 from PIL import Image
+from pydantic import BaseModel
 
 from academia_mcp.files import get_workspace_dir
+from academia_mcp.llm import ChatMessage, llm_acall
 from academia_mcp.settings import settings
-from academia_mcp.llm import llm_acall, ChatMessage
 
+paddlex_logger = logging.getLogger("paddlex")
+paddleocr_logger = logging.getLogger("paddleocr")
+paddlex_logger.setLevel(logging.ERROR)
+paddleocr_logger.setLevel(logging.ERROR)
 
 DESCRIBE_PROMPTS = {
     "general": "Provide a general description of this image. Focus on the main subjects, colors, and overall scene.",
@@ -37,16 +49,64 @@ DESCRIBE_PROMPTS = {
         - If layout is multi-column or tabular, reconstruct lines top-to-bottom, left-to-right; use line breaks between blocks.
         - For any uncertain or low-confidence characters, mark with a '?' and include a note.
         - After the raw extraction, provide a clean, normalized version (fixing obvious OCR artifacts) as a separate section.
-        Return two sections:
+        Return three sections:
+        [GENERAL IMAGE DESCRIPTION]
+        ...
         [RAW TRANSCRIPTION]
         ...
-        [NORMALIZED]
+        [NORMALIZED TRANSCRIPTION]
         ...
         """
     ),
 }
 
 
+class OCRBox(BaseModel):  # type: ignore
+    poly: List[List[float]]
+    text: str
+    score: float
+
+
+class OCRSingleton:
+    instance: Optional[PaddleOCR] = None
+    lock: threading.Lock = threading.Lock()
+
+    @classmethod
+    def get(cls) -> PaddleOCR:
+        if cls.instance is not None:
+            return cls.instance
+        with cls.lock:
+            if cls.instance is None:
+                with open(os.devnull, "w") as devnull:
+                    with contextlib.redirect_stderr(devnull):
+                        cls.instance = PaddleOCR(
+                            use_doc_orientation_classify=False,
+                            use_doc_unwarping=False,
+                            use_textline_orientation=False,
+                        )
+        return cls.instance
+
+
+async def _run_ocr(path: str) -> Dict[str, Any]:
+    def _sync_ocr(path: str) -> Dict[str, Any]:
+        try:
+            ocr = OCRSingleton.get()
+            with open(os.devnull, "w") as devnull:
+                with contextlib.redirect_stderr(devnull):
+                    result = ocr.predict(input=path)[0]
+            rec_texts = result["rec_texts"]
+            rec_scores = result["rec_scores"]
+            rec_polys = result["rec_polys"]
+        except Exception as e:
+            return {"error": str(e)}
+        items = []
+        for poly, text, score in zip(rec_polys, rec_texts, rec_scores):
+            items.append(OCRBox(poly=poly, text=text, score=score).model_dump())
+        return {"boxes": items}
+
+    return await asyncio.to_thread(_sync_ocr, path)
+
+
 def show_image(path: str) -> Dict[str, str]:
     """
     Reads an image from the specified URL or from the current work directory.
@@ -91,7 +151,7 @@ async def describe_image(
             - "general": General description of the image
             - "detailed": Detailed analysis of the image
             - "chess": Analysis of a chess position
-            - "text": Extract and describe text or numbers from the image
+            - "text": Extract and describe text or numbers with an OCR pipeline.
             - "custom": Custom description based on user prompt
     """
     image_base64 = show_image(path)["image_base64"]
@@ -116,4 +176,12 @@ async def describe_image(
         messages=[ChatMessage(role="user", content=content)],
         **llm_kwargs,
     )
+    if description_type == "text":
+        ocr_response = await _run_ocr(path)
+        response = json.dumps(
+            {
+                "vlm_response": response,
+                "ocr_response": ocr_response if ocr_response else [],
+            }
+        )
     return response

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/academia_mcp.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.11.3
+Version: 1.11.4
 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
@@ -12,16 +12,14 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: mcp>=1.10.1
 Requires-Dist: xmltodict>=0.14.0
-Requires-Dist: types-xmltodict>=0.14.0
 Requires-Dist: requests>=2.32.0
-Requires-Dist: types-requests>=2.32.0
 Requires-Dist: pypdf>=5.1.0
 Requires-Dist: beautifulsoup4>=4.12.0
-Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: markdownify==0.14.1
+Requires-Dist: types-xmltodict>=0.14.0
+Requires-Dist: types-requests>=2.32.0
+Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: acl-anthology==0.5.2
-Requires-Dist: markdown==3.7.0
-Requires-Dist: types-markdown==3.7.0.20250322
 Requires-Dist: huggingface-hub>=0.32.4
 Requires-Dist: fire>=0.7.0
 Requires-Dist: openai>=1.97.1
@@ -31,6 +29,8 @@ 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
+Requires-Dist: paddlepaddle>=3.2.0
+Requires-Dist: paddleocr>=3.2.0
 Dynamic: license-file
 
 # Academia MCP
@@ -70,12 +70,16 @@ make install
 ### Quickstart
 - Run over HTTP (default transport):
 ```bash
+python -m academia_mcp --transport streamable-http
+# OR
 uv run -m academia_mcp --transport streamable-http
 ```
 
 - Run over stdio (for local MCP clients like Claude Desktop):
 ```bash
 python -m academia_mcp --transport stdio
+# OR
+uv run -m academia_mcp --transport stdio
 ```
 
 Notes:
@@ -122,7 +126,7 @@ Availability notes:
 - Set one or more of `EXA_API_KEY`, `BRAVE_API_KEY`, `TAVILY_API_KEY` to enable `web_search` and provider tools.
 
 ### Environment variables
-Set as needed depending on which tools you use:
+Set as needed, depending on which tools you use:
 
 - `OPENROUTER_API_KEY`: required for LLM-related tools.
 - `BASE_URL`: override OpenRouter base URL.
@@ -152,7 +156,7 @@ docker run --rm -p 5056:5056 \
   academia_mcp
 ```
 
-Or use existing image: `phoenix120/academia_mcp`
+Or use existing image: [`phoenix120/academia_mcp`](https://hub.docker.com/repository/docker/phoenix120/academia_mcp)
 
 ### Examples
 - [Comprehensive report screencast (YouTube)](https://www.youtube.com/watch?v=4bweqQcN6w8)

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/academia_mcp.egg-info/SOURCES.txt

@@ -25,11 +25,11 @@ academia_mcp/tools/arxiv_search.py
 academia_mcp/tools/bitflip.py
 academia_mcp/tools/document_qa.py
 academia_mcp/tools/hf_datasets_search.py
+academia_mcp/tools/image_processing.py
 academia_mcp/tools/latex.py
 academia_mcp/tools/py.typed
 academia_mcp/tools/review.py
 academia_mcp/tools/s2.py
-academia_mcp/tools/show_image.py
 academia_mcp/tools/speech_to_text.py
 academia_mcp/tools/visit_webpage.py
 academia_mcp/tools/web_search.py
@@ -41,11 +41,11 @@ tests/test_bitflip.py
 tests/test_document_qa.py
 tests/test_extract_json.py
 tests/test_hf_dataset_search.py
+tests/test_image_processing.py
 tests/test_latex.py
 tests/test_review.py
 tests/test_s2.py
 tests/test_server.py
-tests/test_show_image.py
 tests/test_speech_to_text.py
 tests/test_visit_webpage.py
 tests/test_web_search.py

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/academia_mcp.egg-info/requires.txt

@@ -1,15 +1,13 @@
 mcp>=1.10.1
 xmltodict>=0.14.0
-types-xmltodict>=0.14.0
 requests>=2.32.0
-types-requests>=2.32.0
 pypdf>=5.1.0
 beautifulsoup4>=4.12.0
-types-beautifulsoup4>=4.12.0
 markdownify==0.14.1
+types-xmltodict>=0.14.0
+types-requests>=2.32.0
+types-beautifulsoup4>=4.12.0
 acl-anthology==0.5.2
-markdown==3.7.0
-types-markdown==3.7.0.20250322
 huggingface-hub>=0.32.4
 fire>=0.7.0
 openai>=1.97.1
@@ -19,3 +17,5 @@ pymupdf>=1.26.4
 pillow>=11.3.0
 pydantic-settings>=2.6.0
 youtube-transcript-api>=1.2.2
+paddlepaddle>=3.2.0
+paddleocr>=3.2.0

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "academia-mcp"
-version = "1.11.3"
+version = "1.11.4"
 description = "MCP server that provides different tools to search for scientific publications"
 readme = "README.md"
 authors = [
@@ -19,16 +19,14 @@ classifiers = [
 dependencies = [
     "mcp>=1.10.1",
     "xmltodict>=0.14.0",
-    "types-xmltodict>=0.14.0",
     "requests>=2.32.0",
-    "types-requests>=2.32.0",
     "pypdf>=5.1.0",
     "beautifulsoup4>=4.12.0",
-    "types-beautifulsoup4>=4.12.0",
     "markdownify==0.14.1",
+    "types-xmltodict>=0.14.0",
+    "types-requests>=2.32.0",
+    "types-beautifulsoup4>=4.12.0",
     "acl-anthology==0.5.2",
-    "markdown==3.7.0",
-    "types-markdown==3.7.0.20250322",
     "huggingface-hub>=0.32.4",
     "fire>=0.7.0",
     "openai>=1.97.1",
@@ -38,6 +36,8 @@ dependencies = [
     "pillow>=11.3.0",
     "pydantic-settings>=2.6.0",
     "youtube-transcript-api>=1.2.2",
+    "paddlepaddle>=3.2.0",
+    "paddleocr>=3.2.0",
 ]
 
 [dependency-groups]

academia_mcp-1.11.3/tests/test_show_image.py → academia_mcp-1.11.4/tests/test_image_processing.py

@@ -29,3 +29,9 @@ async def test_describe_image_base(test_image_url: str) -> None:
     result = await describe_image(test_image_url)
     assert result is not None
     assert "Interrogator" in result
+
+
+async def test_describe_image_text(test_image_url: str) -> None:
+    result = await describe_image(test_image_url, description_type="text")
+    assert result is not None
+    assert '"text": "Interrogator"' in result

{academia_mcp-1.11.3 → academia_mcp-1.11.4}/tests/test_visit_webpage.py

@@ -7,7 +7,6 @@ def test_visit_webpage_basic() -> None:
     assert content.id == "https://example.com/"
     assert content.provider == "basic"
     assert "Example Domain" in content.text
-    assert "illustrative" in content.text
 
 
 def test_visit_webpage_exa() -> None:
@@ -16,7 +15,6 @@ def test_visit_webpage_exa() -> None:
     assert content.id == "https://example.com/"
     assert content.provider == "exa"
     assert "Example Domain" in content.text
-    assert "illustrative" in content.text
 
 
 def test_visit_webpage_pdf() -> None: