suede-competitor-profiling

Builds evidence-backed competitor profiles from public pages, reviews, and SEO data.

123|10|Updated May 24, 2026
One-click install
npx skills add https://github.com/JasonColapietro/suede-creator-skills --skill suede-competitor-profiling-jasoncolapietro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: suede-competitor-profiling
Source: https://github.com/JasonColapietro/suede-creator-skills/tree/main/skills/suede-competitor-profiling
Command: npx skills add https://github.com/JasonColapietro/suede-creator-skills --skill suede-competitor-profiling-jasoncolapietro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Competitor research often produces unsourced claims, stale pricing, and guesses presented as fact. This Skill turns current public evidence into structured competitor profiles where every field is marked as fact, provider estimate, inference, or unknown, with raw captures saved to disk for audit. ## Core Features & Use Cases - Phased research workflow: Maps and captures public pages (homepage, pricing, features, customers, changelog), optionally collects review and SEO data, then synthesizes profiles with a verification pass before delivery. - Evidence discipline: Persists raw scrapes, SEO exports, and review captures under dated folders, and uses templates with explicit evidence markers so facts, estimates, and inferences stay separate. - Quick scan or deep profile: Defaults to a fast homepage-plus-pricing scan for large competitor lists, with a full deep profile for three or fewer competitors, plus a cross-competitor summary with comparison table and positioning map. - Use Case: Give the agent three competitor URLs and receive one markdown profile per competitor plus a _summary.md with side-by-side metrics, positioning map, and market gaps. ## Quick Start Use suede-competitor-profiling to research https://competitor-a.com and https://competitor-b.com and produce a quick-scan profile for each plus a summary comparison.

Frequently Asked Questions about suede-competitor-profiling

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

FAQPage Schema
How do I profile a competitor's website with an AI agent?

Provide the competitor URLs and the skill maps the site, captures homepage, pricing, features, customers, and changelog pages, then synthesizes a markdown profile per competitor. Raw captures are saved under dated folders so every field can be audited later.

What pages should I scrape for competitor research?

Prioritize homepage, pricing, features, about, customers, integrations, and changelog pages, plus optional review sources like G2, Capterra, Product Hunt, and TrustRadius. Each page type maps to specific fields such as value proposition, tier pricing, and release velocity.

What is the difference between a quick scan and a deep competitor profile?

A quick scan covers homepage and pricing plus one consistent SEO overview, producing an abbreviated profile. A deep profile adds all key pages, reviews, backlink detail, and content strategy analysis, and is the default only for three or fewer competitors.

Does competitor profiling work without Firecrawl or DataForSEO connected?

Yes. The skill inspects tools actually available in the session and falls back to manual browser capture or user-supplied exports. When no SEO provider is available, those fields are marked not collected rather than filled with guesses.

How do I update an existing competitor profile?

Update rather than start fresh: recapture pricing first, refresh SEO metrics only through the same provider and parameters, and scan the changelog. Save new raw data to a new date folder and append a Change Log section noting what changed.

Can I trust competitor traffic numbers in a profile?

Traffic and authority figures are third-party provider estimates, not first-party truth, and are labeled with provider, market, device, and date. The skill forbids presenting estimates, stale pricing, or inferences as verified current fact.