Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A toolkit for scanning documents and text with AI to extract structured entities, then storing results in categorized endpoints on endpoints.work. Results auto-save as JSON files, with optional markdown summaries generated after each run.
Combines AI document extraction and structured storage in a single API call, removing the need to wire together separate OCR, LLM, and database services.
When to use it
- Extracting contact details from business emails
- Parsing invoice PDFs into structured line-item records
- Organizing meeting notes by topic and date
- Tracking job applications across multiple uploaded documents
- Auditing monthly parse and storage usage
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Scan this invoice PDF and add it to my invoice tracker
- 1Call scanFile('./invoice.pdf', 'invoice tracker') against the Endpoints API
- 2API extracts entities such as vendor name, amount, and date from the PDF
- 3Result saves automatically to results/invoice-tracker/ as a timestamped JSON file
- 4Read the saved file and generate a markdown summary in results/summaries/
Structured invoice data stored at /invoice-tracker/{slug}, with extracted entities and a markdown summary report
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.