RugMunchMedia/pryscraper
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
pryscraper/pry_mcp/tools.py
cryptorugmunch f7f9fa0c88 refactor(mcp): split mcp_production.py into pry_mcp/ sub-package 
- pry_mcp/constants.py: protocol constants

- pry_mcp/notifications.py: observer pattern, logging handler, notification helpers

- pry_mcp/tools.py: PRY_TOOLS definitions

- pry_mcp/resources.py: PRY_RESOURCES and PRY_PROMPTS

- pry_mcp/fallback_server.py: FallbackMCPServer implementation

- pry_mcp/sdk_server.py: official MCP SDK path

- mcp_production.py: backward-compatible re-export shim

- Avoid namespace collision with installed mcp SDK by using pry_mcp/
2026-07-03 14:52:59 +02:00

281 lines
11 KiB
Python
Raw Blame History

"""Pry - MCP tool definitions.
PRY_TOOLS list with all scraper templates.
Split from mcp_production.py.
"""
# SPDX-License-Identifier: MIT
# Copyright (c) 2026 Rug Munch Media LLC
#
# Part of Pry - https://git.rugmunch.io/RugMunchMedia/pryscraper
# Licensed under MIT. See LICENSE.
from __future__ import annotations
PRY_TOOLS = [
{
"name": "pry_scrape",
"description": "Scrape a URL to clean markdown. Bypasses Cloudflare, DataDome, Akamai, and other anti-bot systems automatically using a 9-tier fallback (direct → cloudscraper → FlareSolverr → Playwright → Googlebot → Archive.org → Google Cache → Tor → undetected-chromedriver).",
"inputSchema": {
"type": "object",
"properties": {
"url": {"type": "string", "format": "uri", "description": "URL to scrape"},
"bypass_cloudflare": {
"type": "boolean",
"default": True,
"description": "Use FlareSolverr + Playwright fallback",
},
"js_render": {
"type": "boolean",
"default": False,
"description": "Render JavaScript (slower but handles SPAs)",
},
"extract_schema": {
"type": "object",
"description": "Optional JSON schema for structured extraction (e.g., {fields: [{name, selector, type}]})",
},
},
"required": ["url"],
},
"outputSchema": {
"type": "object",
"properties": {
"success": {"type": "boolean"},
"data": {
"type": "object",
"properties": {
"url": {"type": "string"},
"title": {"type": "string"},
"content": {"type": "string", "description": "Clean markdown content"},
"links": {"type": "array", "items": {"type": "string"}},
"images": {"type": "array", "items": {"type": "string"}},
"status_code": {"type": "integer"},
},
},
},
"required": ["success"],
},
},
{
"name": "pry_crawl",
"description": "Crawl a website starting from a URL. Discovers and scrapes linked pages up to max_pages with BFS/DFS traversal.",
"inputSchema": {
"type": "object",
"properties": {
"url": {"type": "string", "format": "uri"},
"max_pages": {
"type": "integer",
"default": 10,
"minimum": 1,
"maximum": 10000,
},
"max_depth": {
"type": "integer",
"default": 2,
"minimum": 0,
"maximum": 20,
},
"include_paths": {
"type": "array",
"items": {"type": "string"},
"description": "URL path patterns to include (e.g., ['/docs/', '/blog/'])",
},
"exclude_paths": {"type": "array", "items": {"type": "string"}},
},
"required": ["url"],
},
"outputSchema": {
"type": "object",
"properties": {
"success": {"type": "boolean"},
"data": {
"type": "object",
"properties": {
"pages": {"type": "array", "items": {"type": "object"}},
"total_pages": {"type": "integer"},
},
},
},
},
},
{
"name": "pry_extract",
"description": "Extract structured data from a URL using CSS selectors. 10x cheaper than LLM extraction. Returns JSON matching the schema.",
"inputSchema": {
"type": "object",
"properties": {
"url": {"type": "string", "format": "uri"},
"schema": {
"type": "object",
"description": "Extraction schema with base_selector and fields array",
"properties": {
"name": {"type": "string"},
"base_selector": {"type": "string"},
"fields": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {"type": "string"},
"selector": {"type": "string"},
"type": {
"type": "string",
"enum": [
"text",
"attribute",
"html",
"nested",
"count",
"exists",
"regex",
],
},
"attribute": {"type": "string"},
"transform": {"type": "string"},
},
"required": ["name", "selector"],
},
},
},
"required": ["name", "fields"],
},
},
"required": ["url", "schema"],
},
"outputSchema": {
"type": "object",
"properties": {
"success": {"type": "boolean"},
"data": {
"type": "object",
"properties": {
"items": {"type": "array", "items": {"type": "object"}},
},
},
},
"required": ["success"],
},
},
{
"name": "pry_template",
"description": "Execute one of 110+ pre-built scraper templates (amazon-product, walmart-product, linkedin-profile, etc.). Returns structured data matching the site's schema.",
"inputSchema": {
"type": "object",
"properties": {
"template_id": {
"type": "string",
"description": "Template ID (e.g., 'amazon-product', 'walmart-product', 'linkedin-profile')",
},
"url": {"type": "string", "format": "uri"},
},
"required": ["template_id", "url"],
},
},
{
"name": "pry_search_templates",
"description": "Search across all 110+ available scraper templates by keyword, category, or site.",
"inputSchema": {
"type": "object",
"properties": {
"q": {
"type": "string",
"description": "Search query (e.g., 'amazon', 'jobs', 'real estate')",
},
"category": {"type": "string", "description": "Filter by category"},
},
"required": ["q"],
},
},
{
"name": "pry_monitor",
"description": "Create a content change monitor. Get notified via webhook, Slack, or Discord when a page changes. AI judges if changes are meaningful.",
"inputSchema": {
"type": "object",
"properties": {
"name": {"type": "string"},
"url": {"type": "string", "format": "uri"},
"schedule_cron": {
"type": "string",
"default": "0 */6 * * *",
"description": "Cron expression (default: every 6 hours)",
},
"webhook_url": {"type": "string", "format": "uri"},
"goal": {
"type": "string",
"description": "Natural language goal for AI change judging (e.g., 'price changes matter, not descriptions')",
},
},
"required": ["name", "url"],
},
},
{
"name": "pry_compliance",
"description": "Check legal compliance for scraping a URL. Analyzes robots.txt, ToS, GDPR/CCPA, and PII exposure. Returns green/yellow/red risk score with recommendations.",
"inputSchema": {
"type": "object",
"properties": {"url": {"type": "string", "format": "uri"}},
"required": ["url"],
},
},
{
"name": "pry_enrich",
"description": "Enrich a URL with company info, tech stack detection, social profiles, and competitive intelligence.",
"inputSchema": {
"type": "object",
"properties": {"url": {"type": "string", "format": "uri"}},
"required": ["url"],
},
},
{
"name": "pry_parse_document",
"description": "Parse a document (PDF, DOCX, image) and extract text/tables. Uses pdfplumber, Tesseract OCR, and python-docx.",
"inputSchema": {
"type": "object",
"properties": {"url": {"type": "string", "format": "uri"}},
"required": ["url"],
},
},
{
"name": "pry_screenshot",
"description": "Take a full-page screenshot of a URL. Returns base64-encoded PNG. Useful for visual analysis and OCR.",
"inputSchema": {
"type": "object",
"properties": {
"url": {"type": "string", "format": "uri"},
"full_page": {"type": "boolean", "default": True},
},
"required": ["url"],
},
},
{
"name": "pry_x402_pricing",
"description": "Get current x402 pay-per-call pricing for all Pry operations. Shows the cost in USDC atomic units for each tool.",
"inputSchema": {"type": "object", "properties": {}, "required": []},
"outputSchema": {
"type": "object",
"properties": {
"success": {"type": "boolean"},
"data": {
"type": "object",
"properties": {
"pricing": {"type": "object"},
"wallet": {"type": "string"},
"facilitator": {"type": "string"},
"supported_networks": {"type": "array", "items": {"type": "string"}},
},
},
},
"required": ["success"],
},
},
{
"name": "pry_referrals",
"description": "Get Pry's referral catalog (60+ providers across LLM, hosting, email, monitoring, proxies, devtools). Each link includes our affiliate ID for revenue attribution.",
"inputSchema": {
"type": "object",
"properties": {"category": {"type": "string"}},
"required": [],
},
},
]
Reference in a new issue View git blame Copy permalink