Loading pattern preview
BlocksReal-world Examples
Competitor Research Agent
Competitor Research Agent
Enter a competitor URL to get a research brief grounded in their own site. Firecrawl maps and scrapes up to four pages, every extracted fact is checked against the scraped text, and an agent adds positioning, gaps, and a sales battle card with quoted evidence.
Use this when you have one competitor URL and need a brief you can trust: pricing, features, and tech read from the site, plus agent analysis that quotes its evidence.
Loading pattern preview
Problems solved
Technical challenges this implementation handles out of the box.
- Extract competitor site data with Firecrawl map and scrape
- Keep model output grounded by verifying it against scraped text
- Analyze positioning and gaps with quoted evidence
- Detect technologies used on competitor properties
- Generate sales battle cards from live competitor pages
Use cases
Products and workflows this pattern is designed to support.
- Sales competitive intelligence
- Pricing strategy research
- Feature gap analysis
- Product roadmap inputs from competitor sites
- Battle cards for account executives
Compare with
Similar patterns, and when each one is the better fit.
SEO AuditUse this when a marketing team needs exportable SEO scorecards that check title, meta description, and Open Graph tags and list top fixes for a page.
Search - Firecrawl (robust)FreeUse this when Firecrawl is your search provider and your chat agent needs to find pages by query rather than scrape URLs you already have.
Firecrawl Landing Page Teardown PDFUse this when you need a conversion teardown PDF comparing your landing page against 2-5 competitor pages, with copy rewrites and a short test plan.
Firecrawl Scrape Tool ChatCrawl and extract content using Firecrawl's API. Includes URL processing, markdown conversion, and structured data extraction.
Setup
Requirements, wiring steps, and what this pattern adds to your project.
Getting started
Pick how you want to pull this pattern in. Then wire env vars and routes the same way.
- 1
Download → Next.js app
Use Download → Next.js app in the preview toolbar. You get a zip scaffold (Pro downloads rewrite preview `/view/.../api/*` paths to `/api/*` automatically).
- 2
Extract the zip and install dependencies
Unzip the archive, then run `pnpm install` (or npm / yarn) in the project folder.
- 3
Fill .env.local
The download includes .env.local placeholders when needed. Add real values for: OPENAI_API_KEY, FIRECRAWL_API_KEY.
- 4
Customize the agent and tool files
Adapt prompts, tools, and stop conditions in the generated project for your product.
- 5
Run the app and open the pattern route
Run `pnpm dev`, open the setup/checklist page if present, then navigate to the generated pattern route and verify responses.
Environment variables
Capabilities
AI SDK APIs
useChatgatewayToolLoopAgenttoolhasToolCallisStepCountgenerateObjecttoolsContextcreateAgentUIStreamResponseInferAgentUIMessageDefaultChatTransportUIMessageProviders
OpenAI
External services
FirecrawlUpstash
Files added
15 files
app/page.tsxapp/layout.tsxapp/api/competitor-agent/route.tslib/actions.tslib/competitor-types.tslib/detect-tech.tslib/grounding.tslib/rate-limit.tslib/research-context.tslib/research-site.tslib/ai/competitor-agent.tslib/ai/tools/analyze-positioning.tslib/ai/tools/find-gaps.tslib/ai/tools/generate-battlecard.tsREADME.md
Dependencies
10 total
npm packages
8AISDK@mendable/firecrawl-jsUpstashlucide-reactNext.jsReactsonnerZod
Registry components
buttoninput
Critical files
8 files
competitor-agent/route.tsAPI route that builds the agent per request and injects the scraped pages as tool context.
ai/competitor-agent.tsToolLoopAgent factory with 3 grounded analysis tools and toolsContext injection.
lib/research-site.tsFirecrawl pipeline: map the site, scrape up to four pages, extract a profile with generateObject, then drop every value the scraped text does not back up.
lib/grounding.tsText checks that keep model output tied to the scraped pages: phrases, numbers, and quotes.
lib/detect-tech.tsSignature-based technology detection from the home page's raw HTML.
tools/analyze-positioning.tsTool that reads positioning from the scraped pages and verifies each quote.
tools/find-gaps.tsTool that finds stated limits (verified by quote) and unmentioned expectations (verified absent).
tools/generate-battlecard.tsTool that writes a battle card about the competitor without inventing facts about the seller.
Requirements
- Complexity
- Advanced
- Setup time
- ~10 minutes
- Node
- 24+
- Memory
- 256MB
Upstash Redis
