docslight 0.1.1__tar.gz → 0.1.3__tar.gz

{docslight-0.1.1 → docslight-0.1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docslight
-Version: 0.1.1
+Version: 0.1.3
 Summary: Lightweight ComPDF document parsing and extraction SDK
 Author-email: ComPDF AI <support@compdf.com>
 License-Expression: MIT
@@ -78,6 +78,7 @@ Parse any document:
 
 ```bash
 docslight parse invoice.pdf --output invoice.md
+docslight parse invoice.pdf --format zip --output invoice.zip
 ```
 
 Extract specific fields:
@@ -86,14 +87,13 @@ Extract specific fields:
 docslight extract invoice.pdf --fields invoice_number,total_amount
 ```
 
-Launch the local Web UI workbench:
+Launch the local API server:
 
 ```bash
-pip install "docslight[web]"
 docslight web
-# Open http://127.0.0.1:8000
+# Health: http://127.0.0.1:8000/api/health
 
-# Or run the same Web UI directly as a module
+# Or run the same API server directly as a module
 python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 ```
 
@@ -103,7 +103,7 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 - **Parse → Markdown** — Convert PDF, DOCX, PPTX, XLSX, and images (PNG, JPG, TIFF, BMP, WebP) to clean Markdown
 - **Extract → JSON** — Pull structured data by field list, JSON Schema, or structured template (key-value + table extraction)
 - **CLI first** — Full-featured command-line interface, script-friendly
-- **Web UI** — Local Flask workbench with drag-and-drop, live preview with bbox highlights, and a Fields Builder UI
+- **API server** — Local Flask backend exposing parse, extract, preview, health, and system-info endpoints
 - **Batch processing** — `parse_batch()` / `extract_batch()` for multiple files
 - **Local LLM extraction** — Ollama or any OpenAI-compatible provider for offline extraction
 - **Document types** — Classify and route documents by type for cloud extraction
@@ -116,7 +116,7 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 | Core SDK & CLI | `pip install docslight` |
 | + Local parsing (OCR, Office) | `pip install "docslight[local]"` |
 | + Local LLM extraction | `pip install "docslight[local,local-llm]"` |
-| + Web UI workbench | `pip install "docslight[web]"` |
+| + API server | `pip install "docslight[web]"` |
 
 > Local CPU parsing is experimental. Validate accuracy and latency on your own documents before production use.
 
@@ -216,35 +216,36 @@ for r in results:
 
 ```bash
 # Parse
-docslight parse invoice.pdf --mode cloud -o invoice.md
-docslight parse invoice.pdf --mode local -o invoice.md
+docslight parse invoice.pdf --mode cloud -o invoice.zip
+docslight parse invoice.pdf --mode cloud --format zip -o invoice.zip
+docslight parse invoice.pdf --mode local -o invoice.zip
 
 # Extract
 docslight extract invoice.pdf --mode cloud --fields invoice_number,total_amount
 docslight extract invoice.pdf --mode local --fields invoice_number --local-llm-provider ollama --local-llm-model llama3.1
+docslight extract "D:\pdf\invoice\1.pdf" --mode local --fields invoice_number --local-llm-provider ollama
 
 # Extract with schema
 docslight extract invoice.pdf --schema schema.json
 
-# Web UI
+# API server
 docslight web --host 127.0.0.1 --port 8000
 ```
 
-## Web UI Workbench
+## API Server
 
-DocSlight Workbench is a local Flask app for visual document processing.
+DocSlight includes a local Flask API server for document processing. Frontend assets are not bundled in this package.
 
 ```bash
-pip install "docslight[web]"
 docslight web
 python -m docslight.web_app
 ```
 
-- **Parse & Extract tabs** — Switch between parsing and extraction workflows
-- **Drag-and-drop upload** — PDF, images, DOCX, PPTX, XLSX
-- **Live preview** — PDF page rendering with bbox highlight overlays
-- **Fields Builder** — Structured UI for building key-value and table extraction templates
-- **Download results** — One-click download of Markdown or JSON output
+- `GET /api/health`
+- `GET /api/system-info`
+- `POST /api/parse`
+- `POST /api/extract`
+- `POST /api/preview`
 
 ## Environment Variables
 

{docslight-0.1.1 → docslight-0.1.3}/README.md

@@ -1,14 +1,14 @@
 <p align="center">
-  <h1 align="center">DocSlight</h1>
+  <h1 align="center">DocSlight</h1>
   <p align="center">Lightweight Python SDK & CLI for document parsing and structured extraction</p>
   <p align="center">
-    <a href="https://pypi.org/project/docslight/"><img src="https://img.shields.io/pypi/v/docslight" alt="PyPI"></a>
-    <a href="https://pypi.org/project/docslight/"><img src="https://img.shields.io/pypi/pyversions/docslight" alt="Python versions"></a>
-    <a href="LICENSE"><img src="https://img.shields.io/github/license/kdanmobile/docslight" alt="License"></a>
+    <a href="https://pypi.org/project/docslight/"><img src="https://img.shields.io/pypi/v/docslight" alt="PyPI"></a>
+    <a href="https://pypi.org/project/docslight/"><img src="https://img.shields.io/pypi/pyversions/docslight" alt="Python versions"></a>
+    <a href="LICENSE"><img src="https://img.shields.io/github/license/kdanmobile/docslight" alt="License"></a>
   </p>
 </p>
 
-## What is DocSlight?
+## What is DocSlight?
 
 A lightweight Python library that turns PDFs, images, and Office documents into clean Markdown or structured JSON — with one line of code. Works with ComPDF Cloud (recommended) or fully offline with local parsers.
 
@@ -23,13 +23,14 @@ print(result.to_markdown())
 ## Quick Start
 
 ```bash
-pip install docslight
+pip install docslight
 ```
 
 Parse any document:
 
 ```bash
 docslight parse invoice.pdf --output invoice.md
+docslight parse invoice.pdf --format zip --output invoice.zip
 ```
 
 Extract specific fields:
@@ -38,16 +39,15 @@ Extract specific fields:
 docslight extract invoice.pdf --fields invoice_number,total_amount
 ```
 
-Launch the local Web UI workbench:
-
-```bash
-pip install "docslight[web]"
-docslight web
-# Open http://127.0.0.1:8000
-
-# Or run the same Web UI directly as a module
-python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
-```
+Launch the local API server:
+
+```bash
+docslight web
+# Health: http://127.0.0.1:8000/api/health
+
+# Or run the same API server directly as a module
+python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
+```
 
 ## Features
 
@@ -55,7 +55,7 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 - **Parse → Markdown** — Convert PDF, DOCX, PPTX, XLSX, and images (PNG, JPG, TIFF, BMP, WebP) to clean Markdown
 - **Extract → JSON** — Pull structured data by field list, JSON Schema, or structured template (key-value + table extraction)
 - **CLI first** — Full-featured command-line interface, script-friendly
-- **Web UI** — Local Flask workbench with drag-and-drop, live preview with bbox highlights, and a Fields Builder UI
+- **API server** — Local Flask backend exposing parse, extract, preview, health, and system-info endpoints
 - **Batch processing** — `parse_batch()` / `extract_batch()` for multiple files
 - **Local LLM extraction** — Ollama or any OpenAI-compatible provider for offline extraction
 - **Document types** — Classify and route documents by type for cloud extraction
@@ -65,10 +65,10 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 
 | Scenario | Command |
 |----------|---------|
-| Core SDK & CLI | `pip install docslight` |
-| + Local parsing (OCR, Office) | `pip install "docslight[local]"` |
-| + Local LLM extraction | `pip install "docslight[local,local-llm]"` |
-| + Web UI workbench | `pip install "docslight[web]"` |
+| Core SDK & CLI | `pip install docslight` |
+| + Local parsing (OCR, Office) | `pip install "docslight[local]"` |
+| + Local LLM extraction | `pip install "docslight[local,local-llm]"` |
+| + API server | `pip install "docslight[web]"` |
 
 > Local CPU parsing is experimental. Validate accuracy and latency on your own documents before production use.
 
@@ -159,44 +159,45 @@ client = DocSlight(
 ### Batch Processing
 
 ```python
-results = client.parse_batch(["doc1.pdf", "doc2.pdf", "doc3.pdf"])
-for r in results:
-    print(r.to_markdown()[:200])
+results = client.parse_batch(["doc1.pdf", "doc2.pdf", "doc3.pdf"])
+for r in results:
+    print(r.to_markdown()[:200])
 ```
 
 ## CLI Usage
 
 ```bash
 # Parse
-docslight parse invoice.pdf --mode cloud -o invoice.md
-docslight parse invoice.pdf --mode local -o invoice.md
+docslight parse invoice.pdf --mode cloud -o invoice.zip
+docslight parse invoice.pdf --mode cloud --format zip -o invoice.zip
+docslight parse invoice.pdf --mode local -o invoice.zip
 
 # Extract
 docslight extract invoice.pdf --mode cloud --fields invoice_number,total_amount
 docslight extract invoice.pdf --mode local --fields invoice_number --local-llm-provider ollama --local-llm-model llama3.1
+docslight extract "D:\pdf\invoice\1.pdf" --mode local --fields invoice_number --local-llm-provider ollama
 
 # Extract with schema
 docslight extract invoice.pdf --schema schema.json
 
-# Web UI
-docslight web --host 127.0.0.1 --port 8000
-```
-
-## Web UI Workbench
-
-DocSlight Workbench is a local Flask app for visual document processing.
-
-```bash
-pip install "docslight[web]"
-docslight web
-python -m docslight.web_app
-```
-
-- **Parse & Extract tabs** — Switch between parsing and extraction workflows
-- **Drag-and-drop upload** — PDF, images, DOCX, PPTX, XLSX
-- **Live preview** — PDF page rendering with bbox highlight overlays
-- **Fields Builder** — Structured UI for building key-value and table extraction templates
-- **Download results** — One-click download of Markdown or JSON output
+# API server
+docslight web --host 127.0.0.1 --port 8000
+```
+
+## API Server
+
+DocSlight includes a local Flask API server for document processing. Frontend assets are not bundled in this package.
+
+```bash
+docslight web
+python -m docslight.web_app
+```
+
+- `GET /api/health`
+- `GET /api/system-info`
+- `POST /api/parse`
+- `POST /api/extract`
+- `POST /api/preview`
 
 ## Environment Variables
 

{docslight-0.1.1 → docslight-0.1.3}/docslight/__init__.py

@@ -19,7 +19,7 @@ from docslight.exceptions import (
 )
 from docslight.result import ExtractResult, ParseResult
 
-__version__ = "0.1.1"
+__version__ = "0.1.2"
 
 __all__ = [
     "AuthenticationError",

{docslight-0.1.1 → docslight-0.1.3}/docslight/cli.py

@@ -81,20 +81,34 @@ def _client_from_args(args: argparse.Namespace) -> DocSlight:
     )
 
 
-def _write_output(content: str, output_path: str | None) -> None:
-    if output_path is None:
-        sys.stdout.write(content)
-        if not content.endswith("\n"):
-            sys.stdout.write("\n")
-        return
-    Path(output_path).write_text(content, encoding="utf-8")
+def _write_output(content: str, output_path: str | None) -> None:
+    if output_path is None:
+        sys.stdout.write(content)
+        if not content.endswith("\n"):
+            sys.stdout.write("\n")
+        return
+    path = Path(output_path)
+    path.write_text(content, encoding="utf-8")
+    sys.stderr.write(f"Wrote {path.resolve()}\n")
+
+
+def _write_binary_output(content: bytes, output_path: str | None) -> None:
+    if output_path is None:
+        output_buffer = getattr(sys.stdout, "buffer", None)
+        if output_buffer is None:
+            raise CLIUsageError("binary output requires --output")
+        output_buffer.write(content)
+        return
+    path = Path(output_path)
+    path.write_bytes(content)
+    sys.stderr.write(f"Wrote {path.resolve()}\n")
 
 
 def _to_pretty_json(data: Any) -> str:
     return json.dumps(data, ensure_ascii=False, indent=2)
 
 
-def run_web_app(host: str, port: int, debug: bool) -> None:
+def run_web_app(host: str, port: int, debug: bool) -> None:
     """Run the optional Flask web application."""
     if importlib.util.find_spec("docslight.web_app") is None:
         raise CLIUsageError(WEB_EXTRA_ERROR)
@@ -105,8 +119,8 @@ def run_web_app(host: str, port: int, debug: bool) -> None:
         if exc.name in {"flask", "werkzeug"}:
             raise CLIUsageError(WEB_EXTRA_ERROR) from exc
         raise
-    _run_web_app = web_app.run_web_app
-    _run_web_app(host, port, debug)
+    _run_web_app = web_app.run_web_app
+    _run_web_app(host, port, debug)
 
 
 def _print_cli_error(error: Exception) -> int:
@@ -125,11 +139,11 @@ def build_parser() -> argparse.ArgumentParser:
     parse_parser = subparsers.add_parser("parse", help="Parse a document")
     parse_parser.add_argument("input")
     parse_parser.add_argument("--output", "-o")
-    parse_parser.add_argument(
-        "--format",
-        choices=("markdown", "json", "standard-json"),
-        default="markdown",
-    )
+    parse_parser.add_argument(
+        "--format",
+        choices=("markdown", "json", "standard-json", "zip"),
+        default=None,
+    )
     _add_common_options(parse_parser)
     parse_parser.set_defaults(func=_run_parse)
 
@@ -151,25 +165,41 @@ def build_parser() -> argparse.ArgumentParser:
     extract_parser.set_defaults(func=_run_extract)
 
     web_parser = subparsers.add_parser("web", help="Run the web application")
-    web_parser.add_argument("--host", default="127.0.0.1")
-    web_parser.add_argument("--port", type=int, default=8000)
-    web_parser.add_argument("--debug", action="store_true")
-    web_parser.set_defaults(func=_run_web)
+    web_parser.add_argument("--host", default="127.0.0.1")
+    web_parser.add_argument("--port", type=int, default=8000)
+    web_parser.add_argument("--debug", action="store_true")
+    web_parser.set_defaults(func=_run_web)
 
     return parser
 
 
-def _run_parse(args: argparse.Namespace) -> int:
-    parse_output = "json" if args.format == "standard-json" else args.format
-    result = _client_from_args(args).parse(args.input, output=parse_output)
-    if args.format == "json":
-        content = _to_pretty_json(result.to_json())
-    elif args.format == "standard-json":
-        content = _to_pretty_json(result.to_standard_json())
-    else:
-        content = result.to_markdown()
-    _write_output(content, args.output)
-    return 0
+def _run_parse(args: argparse.Namespace) -> int:
+    parse_format = _resolve_parse_format(args.format, args.output)
+    parse_output = "json" if parse_format == "standard-json" else "markdown"
+    result = _client_from_args(args).parse(args.input, output=parse_output)
+    if parse_format == "zip":
+        raw_archive = getattr(result, "raw_archive", None)
+        if not isinstance(raw_archive, bytes):
+            raise CLIUsageError("parse result did not include a ZIP archive")
+        _write_binary_output(raw_archive, args.output)
+    elif parse_format == "json":
+        content = _to_pretty_json(result.to_json())
+        _write_output(content, args.output)
+    elif parse_format == "standard-json":
+        content = _to_pretty_json(result.to_standard_json())
+        _write_output(content, args.output)
+    else:
+        content = result.to_markdown()
+        _write_output(content, args.output)
+    return 0
+
+
+def _resolve_parse_format(parse_format: str | None, output_path: str | None) -> str:
+    if parse_format is not None:
+        return parse_format
+    if output_path is not None and Path(output_path).suffix.lower() == ".zip":
+        return "zip"
+    return "markdown"
 
 
 def _run_convert_parse_json(args: argparse.Namespace) -> int:
@@ -197,9 +227,9 @@ def _run_extract(args: argparse.Namespace) -> int:
     return 0
 
 
-def _run_web(args: argparse.Namespace) -> int:
-    run_web_app(args.host, args.port, args.debug)
-    return 0
+def _run_web(args: argparse.Namespace) -> int:
+    run_web_app(args.host, args.port, args.debug)
+    return 0
 
 
 def main(argv: Sequence[str] | None = None) -> int:

{docslight-0.1.1 → docslight-0.1.3}/docslight/cloud/client.py

@@ -171,11 +171,11 @@ class CloudClient:
             return _read_downloaded_result_payload(content)
         return self._response_json(response), None
 
-    def _prepare_options(self, operation: str, options: dict[str, Any]) -> dict[str, Any]:
-        if operation != "extract" or not self._uses_custom_operation_urls():
-            return options
-
-        prepared = dict(options)
+    def _prepare_options(self, operation: str, options: dict[str, Any]) -> dict[str, Any]:
+        if operation != "extract":
+            return options
+
+        prepared = dict(options)
         if "extract_fields" in prepared:
             return prepared
 
@@ -219,7 +219,7 @@ class CloudClient:
         return compacted
 
     def _headers(self) -> dict[str, str]:
-        headers = {"User-Agent": "docslight/0.1.1"}
+        headers = {"User-Agent": "docslight/0.1.2"}
         if self.api_key:
             headers["Authorization"] = f"Bearer {self.api_key}"
             headers["x-api-key"] = self.api_key

{docslight-0.1.1 → docslight-0.1.3}/docslight/preview.py

@@ -1,4 +1,4 @@
-"""Preview rendering helpers for the local Web UI."""
+"""Preview rendering helpers for the local API server."""
 
 from __future__ import annotations
 

{docslight-0.1.1 → docslight-0.1.3}/docslight/schemas/fields.py

@@ -11,8 +11,8 @@ NormalizedFields = list[str] | StructuredFields | None
 ExtractSchema = dict[str, Any]
 
 
-def normalize_fields(fields: list[str] | str | StructuredFields | None) -> NormalizedFields:
-    """Normalize extraction fields from SDK, CLI, or Web UI inputs."""
+def normalize_fields(fields: list[str] | str | StructuredFields | None) -> NormalizedFields:
+    """Normalize extraction fields from SDK, CLI, or API inputs."""
     if fields is None:
         return None
     if isinstance(fields, str):

{docslight-0.1.1 → docslight-0.1.3}/docslight/web_app.py

@@ -2,19 +2,19 @@
 
 from __future__ import annotations
 
-import argparse
-import base64
-import json
-import logging
-import sys
-import tempfile
+import argparse
+import base64
+import json
+import logging
+import sys
+import tempfile
 from collections.abc import Callable
 from io import BytesIO
 from json import JSONDecodeError
 from pathlib import Path
 from typing import Any, cast
 
-from flask import Flask, Response, jsonify, redirect, render_template, request, send_file, url_for
+from flask import Flask, Response, jsonify, request, send_file
 from werkzeug.datastructures import FileStorage
 from werkzeug.utils import secure_filename
 
@@ -58,21 +58,18 @@ OFFICE_PREVIEW_UNSUPPORTED_MESSAGE = (
 LOG_FORMAT = "%(levelname)s:%(name)s:%(message)s"
 
 
-def create_app(docslight_factory: Callable[..., Any] = DocSlight) -> Flask:
-    """Create the local DocSlight Flask application."""
-    app = Flask(__name__)
-
-    @app.get("/")
-    def index() -> Any:
-        return redirect(url_for("parse_page"))
-
-    @app.get("/parse")
-    def parse_page() -> str:
-        return render_template("parse.html", active_page="parse")
-
-    @app.get("/extract")
-    def extract_page() -> str:
-        return render_template("extract.html", active_page="extract")
+def create_app(docslight_factory: Callable[..., Any] = DocSlight) -> Flask:
+    """Create the local DocSlight Flask application."""
+    app = Flask(__name__)
+
+    @app.get("/")
+    def index() -> Any:
+        return jsonify(
+            {
+                "status": "healthy",
+                "service": "docslight-web",
+            }
+        )
 
     @app.get("/api/health")
     def health() -> Any:
@@ -160,7 +157,11 @@ def create_app(docslight_factory: Callable[..., Any] = DocSlight) -> Flask:
     return app
 
 
-def run_web_app(host: str = "127.0.0.1", port: int = 8000, debug: bool = False) -> None:
+def run_web_app(
+    host: str = "127.0.0.1",
+    port: int = 8000,
+    debug: bool = False,
+) -> None:
     """Run the local DocSlight web application."""
     _configure_web_logging(debug)
     create_app().run(host=host, port=port, debug=debug)
@@ -183,17 +184,17 @@ def build_parser() -> argparse.ArgumentParser:
         prog="python -m docslight.web_app",
         description="Run the DocSlight web application.",
     )
-    parser.add_argument("--host", default="127.0.0.1")
-    parser.add_argument("--port", type=int, default=8000)
-    parser.add_argument("--debug", action="store_true")
-    return parser
+    parser.add_argument("--host", default="127.0.0.1")
+    parser.add_argument("--port", type=int, default=8000)
+    parser.add_argument("--debug", action="store_true")
+    return parser
 
 
 def main(argv: list[str] | None = None) -> int:
-    """Run the standalone DocSlight web application entrypoint."""
-    args = build_parser().parse_args(argv)
-    run_web_app(args.host, args.port, args.debug)
-    return 0
+    """Run the standalone DocSlight web application entrypoint."""
+    args = build_parser().parse_args(argv)
+    run_web_app(args.host, args.port, args.debug)
+    return 0
 
 
 def local_llm_from_form(form: Any) -> dict[str, str] | None:

{docslight-0.1.1 → docslight-0.1.3}/docslight.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docslight
-Version: 0.1.1
+Version: 0.1.3
 Summary: Lightweight ComPDF document parsing and extraction SDK
 Author-email: ComPDF AI <support@compdf.com>
 License-Expression: MIT
@@ -78,6 +78,7 @@ Parse any document:
 
 ```bash
 docslight parse invoice.pdf --output invoice.md
+docslight parse invoice.pdf --format zip --output invoice.zip
 ```
 
 Extract specific fields:
@@ -86,14 +87,13 @@ Extract specific fields:
 docslight extract invoice.pdf --fields invoice_number,total_amount
 ```
 
-Launch the local Web UI workbench:
+Launch the local API server:
 
 ```bash
-pip install "docslight[web]"
 docslight web
-# Open http://127.0.0.1:8000
+# Health: http://127.0.0.1:8000/api/health
 
-# Or run the same Web UI directly as a module
+# Or run the same API server directly as a module
 python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 ```
 
@@ -103,7 +103,7 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 - **Parse → Markdown** — Convert PDF, DOCX, PPTX, XLSX, and images (PNG, JPG, TIFF, BMP, WebP) to clean Markdown
 - **Extract → JSON** — Pull structured data by field list, JSON Schema, or structured template (key-value + table extraction)
 - **CLI first** — Full-featured command-line interface, script-friendly
-- **Web UI** — Local Flask workbench with drag-and-drop, live preview with bbox highlights, and a Fields Builder UI
+- **API server** — Local Flask backend exposing parse, extract, preview, health, and system-info endpoints
 - **Batch processing** — `parse_batch()` / `extract_batch()` for multiple files
 - **Local LLM extraction** — Ollama or any OpenAI-compatible provider for offline extraction
 - **Document types** — Classify and route documents by type for cloud extraction
@@ -116,7 +116,7 @@ python -m docslight.web_app --host 0.0.0.0 --port 8000 --debug
 | Core SDK & CLI | `pip install docslight` |
 | + Local parsing (OCR, Office) | `pip install "docslight[local]"` |
 | + Local LLM extraction | `pip install "docslight[local,local-llm]"` |
-| + Web UI workbench | `pip install "docslight[web]"` |
+| + API server | `pip install "docslight[web]"` |
 
 > Local CPU parsing is experimental. Validate accuracy and latency on your own documents before production use.
 
@@ -216,35 +216,36 @@ for r in results:
 
 ```bash
 # Parse
-docslight parse invoice.pdf --mode cloud -o invoice.md
-docslight parse invoice.pdf --mode local -o invoice.md
+docslight parse invoice.pdf --mode cloud -o invoice.zip
+docslight parse invoice.pdf --mode cloud --format zip -o invoice.zip
+docslight parse invoice.pdf --mode local -o invoice.zip
 
 # Extract
 docslight extract invoice.pdf --mode cloud --fields invoice_number,total_amount
 docslight extract invoice.pdf --mode local --fields invoice_number --local-llm-provider ollama --local-llm-model llama3.1
+docslight extract "D:\pdf\invoice\1.pdf" --mode local --fields invoice_number --local-llm-provider ollama
 
 # Extract with schema
 docslight extract invoice.pdf --schema schema.json
 
-# Web UI
+# API server
 docslight web --host 127.0.0.1 --port 8000
 ```
 
-## Web UI Workbench
+## API Server
 
-DocSlight Workbench is a local Flask app for visual document processing.
+DocSlight includes a local Flask API server for document processing. Frontend assets are not bundled in this package.
 
 ```bash
-pip install "docslight[web]"
 docslight web
 python -m docslight.web_app
 ```
 
-- **Parse & Extract tabs** — Switch between parsing and extraction workflows
-- **Drag-and-drop upload** — PDF, images, DOCX, PPTX, XLSX
-- **Live preview** — PDF page rendering with bbox highlight overlays
-- **Fields Builder** — Structured UI for building key-value and table extraction templates
-- **Download results** — One-click download of Markdown or JSON output
+- `GET /api/health`
+- `GET /api/system-info`
+- `POST /api/parse`
+- `POST /api/extract`
+- `POST /api/preview`
 
 ## Environment Variables