competitor-intel

Scrapes competitor pricing pages with Playwright and extracts structured feature intelligence.

Updated Mar 2, 2026
One-click install
npx skills add https://github.com/LeroyAdonis/kitfix-2.0 --skill competitor-intel-leroyadonis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: competitor-intel
Source: https://github.com/LeroyAdonis/kitfix-2.0/tree/main/.github/skills/competitor-intel
Command: npx skills add https://github.com/LeroyAdonis/kitfix-2.0 --skill competitor-intel-leroyadonis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, playwright-extra, puppeteer-extra-plugin-stealth.

What problem does it solve? Manually researching competitor pricing pages and feature lists is slow and inconsistent. This Skill automates the collection of competitor intelligence from public marketing pages and turns it into structured, comparable data for product decisions. ## Core Features & Use Cases - Automated Web Scraping: Uses Playwright to visit competitor pricing and marketing pages, capture full-page screenshots, and extract page text. - Structured Extraction: Uses Gemini to parse page content into a typed JSON schema covering pricing tiers, AI features, supported platforms, languages, and trial offers. - Gap Analysis & Kanban Output: Compares competitor capabilities against your own product and generates prioritized Kanban card suggestions stored in a database. - Use Case: A product manager wants to know how Buffer, Hootsuite, and Later price their plans and which AI features they offer. The Skill scrapes all three pages, persists records to the database, and returns a report with feature gap suggestions. ## Quick Start Research the pricing pages of Buffer, Hootsuite, and Later and give me a feature gap report with Kanban suggestions.

Frequently Asked Questions about competitor-intel

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I scrape competitor pricing pages with Playwright?▼

Launch a headless Chromium browser with Playwright, navigate to the pricing URL with networkidle wait, take a full-page screenshot, and extract document.body.innerText. The extracted text is then classified into structured fields like pricing tiers and features.

How to extract structured data from scraped web page text?▼

Send the extracted page text to Gemini with a prompt describing the target TypeScript interface and instruct it to return only valid JSON. Slice the text to around 8000 characters to stay within context limits and retry with a stricter prompt if parsing fails.

Can this scrape pages that require login or sign-up?▼

No. The Skill is explicitly restricted to public marketing and pricing pages only. It never attempts login, sign-up, or trial flows, and never creates accounts on competitor services.

Why does Playwright page scraping time out or return a blank page?▼

Timeouts usually come from networkidle never settling; increase the timeout to 60 seconds or switch to domcontentloaded. Blank pages often indicate bot detection, which can be mitigated with playwright-extra and a stealth plugin, confirmed via screenshot.

How many competitor websites can be analyzed in one session?▼

The Skill caps each session at 5 competitor URLs to prevent context bloat. Each URL produces a screenshot, a structured database record, and contributes to the final gap analysis report.