Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Remove accidentally included test script
What This Skill Does
Shelv converts PDF documents into structured Markdown filesystems accessible via standard Unix tools. Upload a PDF through the API, wait for async processing, then extract the result into your workspace as real files. Supports templates for books, legal contracts, and academic papers.
Converts opaque PDFs into grep-able, find-able Markdown files without any custom parsing code or PDF libraries.
When to use it
- Searching a legal contract for specific clauses with grep
- Extracting chapters from a research paper into browsable files
- Converting a regulatory document for version-controlled editing
- Scanning a book's structure before reading
- Pulling targeted sections from a PDF without downloading the full archive
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: convert contract.pdf into searchable files and find all force majeure clauses
- 1Upload contract.pdf using shelv-upload.sh with --template legal-contract and capture the shelf ID
- 2Poll processing status with shelv-poll-status.sh until shelf reaches ready state
- 3Hydrate the shelf into the workspace using shelv-hydrate.sh
- 4Run grep -r "force majeure" over the hydrated directory to find matching files
- 5Read matching clause files with cat
Markdown files for each contract section under ~/.openclaw/workspace/shelves/contract/, with force majeure clause content printed to terminal
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.