Pdf2word Skills

Convert scanned PDF documents into Word text documents using a free, local OCR engine or remote api.

Security Scan

The skill's files and instructions are coherent with its stated purpose (download a local OCR binary, run it on a PDF, and convert extracted text into a .docx); no broad unexplained permissions or hidden exfiltration in the provided code are present, but you should vet the downloaded binary and be cautious about sending sensitive documents to remote OCR engines.

Purpose & Capability

The name/description match the delivered assets: a Python script that calls a local 'docr' binary and uses python-docx to produce a .docx. The included install.sh downloads the expected OCR binary from a GitHub releases URL — this is consistent with providing a local OCR engine.

Instruction Scope

SKILL.md stays on task (install binary, pip deps, run script). It also documents optional use of remote engines (e.g., Gemini) and instructs creating ~/.ocr/config with a gemini_api_key. That config step is outside the skill directory and is not declared in required env/config fields; it's optional but relevant to user privacy and should be noted.

Install Mechanism

The install script downloads a single binary from a GitHub releases URL and writes it under scripts/docr/. Downloading from GitHub releases is a typical, low-risk mechanism compared with arbitrary IPs or paste sites. The script does not extract archives or run additional installers. However, the binary will be executed, so its provenance should be validated.

Credentials

No required environment variables are declared, and the Python script does not read secrets itself. However, SKILL.md asks users to store API keys in ~/.ocr/config for optional remote engines (Gemini). That is reasonable for optional remote OCR but is not declared in requires.env and should be considered a configuration that affects privacy/security for sensitive docs.

Persistence & Privilege

The skill does not request always:true, does not modify other skills, and only places the downloaded binary under the skill's scripts directory (and optionally asks the user to create ~/.ocr/config). There is no permanent elevated privilege requested.

Assessment

This skill appears to do what it claims: it downloads a docr binary, runs it on PDFs, and builds a .docx from the extracted text. Before installing or running it: 1) Inspect the referenced GitHub repo/releases (https://github.com/scottkiss/doc-ocr) and verify the release and maintainer match your trust criteria; prefer checking a checksum or signed release if available. 2) Run the install and conversion in a sandbox or VM if you will process sensitive documents, because the downloaded binary is third-party native code and could perform network activity. 3) If you plan to use a remote engine (Gemini), understand that text may leave your machine and follow your organization's data-sharing policies; SKILL.md suggests storing the API key in ~/.ocr/config (this is optional but not declared elsewhere). 4) On Windows there may be an executable extension mismatch (install creates docr.exe but the Python script looks for 'docr'); verify behavior on your platform before automating. 5) If you need stronger assurance, request the upstream source code/binary reproducible build or replace the binary with a vetted OCR implementation.

PDF to Word Converter

🇨🇳 简体中文 / Simplified Chinese

A skill to extract text from scanned PDF documents and convert them into reusable Word (.docx) files using the free, local docr OCR engine.

Prerequisites

1. Initialize the OCR engine by downloading the binaries:

   bash scripts/install.sh
   

2. Install the required Python dependencies:

   pip install -r scripts/requirements.txt
   

Usage

Run the Python script passing the input PDF file and the desired output .docx file path. You can also append any additional standard docr arguments (such as engine preferences).

python scripts/pdf2word.py <input.pdf> <output.docx> [docr_args...]

Examples

Convert a single file with the default local engine:

python scripts/pdf2word.py sample.pdf sample_output.docx

Using Other API Engines

By default, the script uses the local RapidOCR engine. The underlying docr tool also supports other engines like the Google Gemini API for potentially higher recognition accuracy on complex layouts.

To use Gemini, first configure your API key:

mkdir -p ~/.ocr
echo "gemini_api_key=your_gemini_key" > ~/.ocr/config

Then pass the -engine gemini argument to the script:

python scripts/pdf2word.py sample.pdf sample_output.docx -engine gemini

If your document has tables, you can force Gemini to output them in Markdown format so the script can parse them into native Word tables:

python scripts/pdf2word.py sample.pdf sample_output.docx -engine gemini -prompt "Extract all text and preserve tables in Markdown format using | symbols."

How it Works

1. The script calls docr, which uses the specified OCR model (RapidOCR by default) to read text from the scanned PDF.
2. The extracted text is temporarily stored.
3. The python-docx library is used to read the temporary text and construct a formatted Word document.
4. Temporary files are cleaned up automatically.