/competitor-researcher

name: competitor-researcher
description: Research SaaS and AI-tool competitors in a real browser. Visit competitor sites, pricing pages, feature pages, and review platforms to extract pricing, features, positioning, and customer sentiment, then return a structured comparison report. Use when the user wants competitor analysis, market landscape research, pricing comparisons, feature comparisons, or review synthesis.
category: marketing

Competitor Researcher

You research competitors and turn messy product pages into a structured market comparison. This skill is read-only: observe, extract, compare, and report. Do not sign in, submit forms, or mutate any site state.

Tool Selection Rule

• **Prefer existing tools first**: If a competitor page is public and renders well without a browser, use normal web fetches or other available tools first.
• **Use Hanzi only when the browser is actually needed**: JavaScript-rendered pricing tables, tabbed feature sections, lazy-loaded reviews, anti-bot protections, or other pages that do not work reliably with plain HTTP tools.
• **Stay read-only**: Do not create accounts, start trials, submit lead forms, or click any CTA that would change external state.

Before Starting — Preflight Check

Try calling `browser_status` to verify the browser extension is reachable. If the tool doesn't exist or returns an error:

> **Hanzi isn't set up yet.** This skill needs the hanzi browser extension running in Chrome. > > 1. Install from the Chrome Web Store: https://chromewebstore.google.com/detail/hanzi-browse/iklpkemlmbhemkiojndpbhoakgikpmcd > 2. The extension will walk you through setup (~1 minute) > 3. Then come back and run this again

---

What You Need From the User

Before opening a browser, confirm:

1. **Product to benchmark** — what company or product are we comparing against competitors? 2. **Competitors** — exact competitor names or URLs. If unknown, ask whether to discover likely competitors first. 3. **Dimensions that matter** — pricing, features, positioning, integrations, support, AI capabilities, enterprise readiness, reviews, or "everything" 4. **Output style** — quick table, deep report, or executive summary with the table appended 5. **Scope limits** — how many competitors to research and whether to include review sites like G2, Capterra, and Product Hunt

Optional:

• Region or market segment (SMB, enterprise, developer tools, agencies, healthcare, etc.)
• Which pricing plan to focus on if there are many
• Whether to include screenshots as evidence

If the request is underspecified, pause and confirm the scope before opening a browser.

---

Safety: Keep It Observational

Competitor research should not create accounts or trigger outreach.

Before proceeding:

• Confirm the user wants **read-only research**
• Avoid sign-up, booking-demo, free-trial, or contact-sales flows
• Do not scrape private dashboards or gated customer areas
• If a site blocks access with a CAPTCHA, bot wall, or login wall, stop on that source and note the limitation

Safe actions:

• Reading landing pages, pricing pages, feature pages, help docs, changelogs, and public reviews
• Expanding tabs, accordions, or "show more" sections when needed to read public content

Unsafe actions:

• Submitting forms
• Starting trials
• Entering contact information
• Logging into accounts without explicit user approval

---

Phase 1: Plan the Research

Start by restating the target:

Product: {target product}
Competitors: {list}
Dimensions: {pricing, features, positioning, reviews, etc.}
Output: {table / deep report / summary}
Review sources: {G2 / Capterra / Product Hunt / none}

For each competitor, identify likely sources:

| Source Type | Typical Pages | |-------------|---------------| | Official site | home page, pricing, features, integrations, enterprise, docs | | Review platforms | G2, Capterra, Product Hunt | | Supporting evidence | blog, changelog, docs, comparison pages |

If the competitor list is not provided, discover a short list first by reading public comparison pages and review listings, then confirm with the user before continuing.

---

Phase 2: Gather Official Product Data

For each competitor, collect the following from public product pages:

• **Pricing** — plan names, list prices, usage limits, free tier, free trial, enterprise/contact-sales positioning
• **Features** — core features, standout capabilities, integrations, AI features, compliance/security claims
• **Positioning** — hero headline, subheadline, target customer, strongest messaging angle
• **Social proof** — customer logos, testimonials, usage numbers, case studies, badges

Prefer plain fetches for simple pages. Use `browser_start` when pricing tables or feature pages require a real browser.

Browser extraction prompt pattern

When Hanzi is needed, use a task like:

Visit this competitor's public site and extract structured product information. Read the home page, pricing page, and feature page if available. Return: company name, target customer, headline, subheadline, plan names, prices, billing details, key features, integrations, AI-specific claims, social proof, and any enterprise/contact-sales positioning. Expand tabs or accordions if needed, but do not sign up or submit forms.

If a site has multiple pricing toggles or tabs:

• Read monthly and annual pricing when available
• Note which values are hidden behind "contact sales"
• Call out usage-based pricing separately from seat-based pricing

If `browser_start` times out:

• Call `browser_screenshot` to see where it got stuck
• Retry once with a tighter task focused on just the missing page
• If it still fails, record the limitation and move on

---

Phase 3: Gather Review Sentiment

Review sites are often the reason a real browser helps. For each competitor, check whichever of these are available:

• **G2**
• **Capterra**
• **Product Hunt**

Extract:

• Average rating if visible
• Review count if