Pdf Parser Agent

Parses local PDF files into structured Markdown and JSON using opendataloader-pdf for deterministic, local document content extraction.

Security Scan

The skill's code, instructions, and requirements are consistent with a local PDF-to-Markdown/JSON parser; nothing requested or installed is disproportionate to that purpose.

Purpose & Capability

Name and description match the included script and SKILL.md: the skill runs a local Python-based converter (opendataloader-pdf) on local PDF files. The only minor oddity is an included package.json (Node metadata) despite this being a Python script; this appears cosmetic and does not contradict the stated purpose.

Instruction Scope

SKILL.md instructs the agent to run a local script against local PDFs and references a single dependency (opendataloader-pdf). The runtime instructions do not request unrelated files, environment variables, or external endpoints. The script does append the user's site-packages path to sys.path to locate a --user pip installation, which is reasonable for a dependency lookup but means it will import whatever opendataloader-pdf is installed in the user's site.

Install Mechanism

No install spec is provided by the skill (instruction-only), so nothing is downloaded or written by the skill itself. Dependency installation is left to the user (pip install --user opendataloader-pdf). This is low-risk for the skill bundle, though the external Python package remains a separate trust decision.

Credentials

The skill declares no environment variables, credentials, or config paths and its code does not read secrets. It only reads a user-supplied local file path (validated to be inside the current workspace) and writes output to a specified directory — which is proportionate to the stated function.

Persistence & Privilege

The skill does not request persistent or elevated presence (always:false). It does not modify other skills or system-wide agent settings. Autonomous invocation is allowed by default but is not combined with any broad credential access or unusual privileges.

Assessment

This skill appears to do what it says: convert local PDFs using the opendataloader-pdf package. Before installing or running it: - Inspect and vet the external dependency (opendataloader-pdf) you will pip-install; that package will execute on your machine and is the primary runtime risk. Prefer to install it in a fresh virtualenv rather than system-wide. - Note the script adds your user-site-packages to sys.path, so whatever is installed there will be imported. If you share an environment, ensure no untrusted packages are present in user-site. - The script validates that input files are inside the current working directory; still run it in a controlled workspace to avoid accidental processing of sensitive files. - The included package.json is unexpected for a Python-only skill but appears harmless; it may be leftover metadata. - If you need stronger isolation, run this tool in a container or VM and audit opendataloader-pdf's behavior (it may spawn Java or other subprocesses according to the tests/notes).

SKILL.md - pdf-parser-agent

Purpose

Parses local PDF files into structured Markdown and JSON formats using the opendataloader-pdf library, providing deterministic, local data extraction that bypasses LLM context limits for document content ingestion.

Core Technology Attribution

This skill is built upon opendataloader-pdf, originally developed by bundolee and claude.

Dependencies

This skill requires Python packages installed system-wide or user-site-wide:

1. opendataloader-pdf

Usage Example

The skill's execution script dynamically finds the correct user-site packages path, assuming the user has installed the dependency via pip install --user opendataloader-pdf.

# Assuming a PDF exists at 'Files for testing/sample-local-pdf.pdf'
openclaw skill pdf-parser-agent --run --args "Files for testing/sample-local-pdf.pdf"

Implementation Notes

The underlying logic now uses site.getusersitepackages() to dynamically locate the installed package, maximizing portability across different OS/Python minor versions.