Stop Guessing What to Build Next
AgentOS Scope Out crawls PropTech supplier directories, scores every product on replicability, demand, revenue potential, and competitive gaps — then hands you a ranked shortlist of the best SaaS opportunities to build.
From Directory URL to Build-Ready Dossier in Minutes
Paste a Directory URL
Drop a PropTech supplier directory URL (starting with Kerfuffle.com) into the crawl trigger. Choose between a quick skim to gauge directory size or a full analysis run. The pipeline starts immediately.
The System Does the Heavy Lifting
Scope Out crawls every listing, scrapes reviews, ratings, pricing signals, and feature lists from each product page. A real-time pipeline status panel shows you exactly how many products have been discovered, scraped, scored, and dossier-generated as it runs.
Review a Ranked Opportunity Table
Every analysed product lands in a sortable, filterable dashboard ranked by composite opportunity score. Sort by replicability, market demand, revenue potential, or competitive gap. Filter by PropTech category or pricing tier. Drill into any row for the full dossier.
Copy the Dossier Straight Into Your Build Workflow
Each product dossier includes an auto-drafted mission statement, a prioritised 10-item feature list, and a competitor weakness analysis — all with one-click copy-to-clipboard in Markdown. Export as PDF or CSV for offline review.
48 Features Built Around One Question: What Should We Build?
Four-Dimension AI Scoring Engine
Every product is scored across replicability (how buildable with your stack), market demand (review volume and sentiment), revenue potential (pricing tier × market size), and competitive gaps (what negative reviews reveal). A weighted composite score ranks the full field.
Automated Directory Crawling
Paste a Kerfuffle URL and the system crawls every listing page, extracts product names, categories, descriptions, and source URLs, and persists them instantly. Handles pagination, rate-limiting, and deduplication automatically.
Ranked Products Dashboard
A data-dense, sortable table — not cards. Sort by any of the four scoring dimensions, filter by PropTech category or pricing tier, search across product names and dossier content, and share filtered views via URL query params.
Build-Ready Product Dossiers
Each dossier contains an auto-drafted mission statement, a suggested 10-item feature list prioritised by market gaps, and a competitor weakness analysis drawn from negative reviews. Every section has a one-click Markdown copy button.
Competitive Gap Mining
The gap scorer reads exclusively negative reviews and low-rating patterns, surfaces the top 3 recurring complaints, and maps them to specific product opportunity themes — 'no mobile app', 'poor API', 'pricing too high' — with frequency and severity scoring.
Score Movement Tracking
When a product's composite score shifts by more than 1.5 points between crawl runs, a directional indicator (▲ / ▼) appears on the dashboard row. Shortlisted products trigger an in-app alert on significant score drops.
Side-by-Side Product Comparison
Select 2–4 products and open a comparison table aligning all four dimension scores, pricing tier, category, review count, and weakness themes in columns. Built for decisions, not browsing.
Scheduled Re-Crawls
Mark any directory as monitored and a weekly cron job re-crawls it automatically. Newly discovered products are badged 'new since last crawl' and trigger an in-app notification — so you never miss a new entrant.
Configurable Score Weighting
Adjust the relative weight of each scoring dimension to match your current build priorities. Change weights and the dashboard re-ranks instantly without re-running AI analysis.
Real-Time Pipeline Status
A live progress panel shows each stage of every crawl run: listings discovered, products deep-scraped, products scored, dossiers generated — with counts and completion states updated in real time.
Product Shortlist & Status Lifecycle
Tag products as new, reviewed, shortlisted, building, or archived. Shortlisted products surface in a dedicated Watchlist view. Status filters keep the ranked table focused on what's actually actionable.
Team Notes on Every Dossier
Any team member can add timestamped, attributed notes to a product dossier — 'too complex', 'flagged for Q2', 'pricing too high to compete'. Notes thread on the dossier page and persist across re-scores.
Internal-Grade Security for an Internal Tool
Scope Out is restricted to your team by design — not by policy.
Email Allowlist Access Gate
Sign-in is restricted to an approved email allowlist. Any attempt from an unlisted address is rejected immediately with a clear 'access restricted' message. No public sign-up flow exists.
Robots.txt & Rate-Limit Compliance
The crawler respects robots.txt rules, enforces per-domain rate limits (default 1 req/sec), and identifies itself with a proper user-agent header. Exponential back-off handles 429 and 503 responses cleanly.
AI Cost Controls
Token usage and estimated OpenAI cost are tracked per crawl run and stored monthly. Configurable per-run cost thresholds trigger an in-app alert before spend accumulates unnoticed.
Scrape Cache & Data Retention Policy
Raw scrape data is retained for a configurable window (default 90 days). A minimum re-scrape interval (default 7 days) prevents redundant API calls and runaway data growth.