An intelligent document processing skill that automatically extracts summaries, answers, and structured entities from any URL, PDF, or raw text input.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install doc-miner
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install doc-miner using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Doc Miner?
Doc Miner is a versatile tool designed to transform unstructured content into actionable intelligence. By leveraging advanced AI protocols, this skill analyzes PDFs, web pages, and raw text to provide context-aware results. It is a vital addition to the library of Openclaw Skills, allowing developers to automate the heavy lifting of document analysis without manual intervention.
The skill is built on a smart orchestration layer that auto-detects the required processing mode based on your specific task. Whether you need a high-level summary, a specific answer to a complex question, or a list of structured data points like dates and currency, Doc Miner handles the routing and extraction logic seamlessly.
Doc Miner Use Cases
- Summarizing extensive research papers, long-form articles, or corporate PDFs.
- Performing targeted Q&A sessions against document contents to find specific facts.
- Extracting structured data such as named entities, important dates, and financial figures for database entry.
- Conducting deep literature reviews and competitive research by processing multiple web sources.
- Analyzing raw text blocks or logs where a source URL is not available.
How Doc Miner Works
- The user provides a source input consisting of either a public URL (webpage or PDF) or direct raw text.
- An optional task description is provided; the system analyzes the intent to determine if it should summarize, answer a question, or extract data.
- The request is routed through the AIProx orchestration endpoint using Openclaw Skills architecture.
- The document is processed transiently to maintain privacy and security.
- The skill returns a structured JSON response containing mode-specific metadata such as key points, confidence scores, or entity lists.
Doc Miner Setup
To use Doc Miner, you must have an active AIProx account and an environment variable set for authentication.
- Obtain your API token from the AIProx dashboard.
- Export the token to your environment:
export AIPROX_SPEND_TOKEN="your_token_here"
- Interface with the skill via a standard POST request:
curl -X POST https://aiprox.dev/api/orchestrate \
-H "Content-Type: application/json" \
-H "X-Spend-Token: $AIPROX_SPEND_TOKEN" \
-d '{
"task": "summarize",
"url": "https://example.com/report.pdf"
}'
Doc Miner Data Schema & Taxonomy
Doc Miner organizes output based on the detected mode. The following table outlines the metadata taxonomy for these Openclaw Skills:
| Mode | Key Fields Returned |
|---|---|
| Summary | summary, key_points, word_count, source_type |
| Q&A | answer, context, confidence, source_type |
| Extraction | entities, dates, numbers, key_points, source_type |
All responses include a mode field to identify which processing logic was applied to the input data.
Doc Miner Advanced Features
- Auto-Mode Detection: Eliminates the need for manual configuration by sniffing the task intent (e.g., 'extract' vs 'what is').
- Multi-Source Support: Seamlessly handles both hosted PDF/HTML URLs and direct string injections.
- Entity Recognition: Automatically identifies and categorizes dates, monetary values, and organizations within the text.
- Transient Processing: Ensures data privacy by analyzing content in real-time without persistent storage.
- Confidence Scoring: Provides context and reliability metrics for Q&A tasks to ensure high-accuracy research.
SKILL.md
Loading
Related Openclaw Skills
Feishu Card Sender Beautify
A specialized tool for creating and sending visually appealing Feishu Interactive Cards for professional notifications.
Polyglot AI Translation Skill
A high-fidelity translation agent that provides cultural nuance and formality control for sophisticated multilingual workflows.
Code Reviewer
An automated code review assistant that analyzes pull requests for security risks, complexity, and architectural integrity across multiple programming languages.
ExpenseLog Conversational Expense Tracking
A conversational AI skill that enables natural language expense logging and real-time budget tracking without manual spreadsheet entry.
Cloud Doc Intelligent Assistant
A specialized data collection tool for scraping, localizing, and monitoring changes across major cloud service provider documentation.
ClawPolicy
ClawPolicy is an explainable autonomous execution policy engine designed for low-touch, auditable agent execution and policy supervision.
Featured*
