deal-pipeline

star 0

Full 6-stage deal intelligence pipeline for finding motivated-seller opportunities across any asset class. Scans CREXI, auctions, BizBuySell, and Zillow, applies regional market intelligence and prediction market signals, detects hidden gems, and produces ranked HOT/WARM/WATCH deals with outreach scripts. Use when user says "run hunt", "find deals", "scan for opportunities", "/hunt", "run the pipeline", "find me deals in Nevada County", "what deals are available today", "run all zones", or "start the deal pipeline".

JeromyJSmith By JeromyJSmith schedule Updated 3/5/2026

name: deal-pipeline description: Full 6-stage deal intelligence pipeline for finding motivated-seller opportunities across any asset class. Scans CREXI, auctions, BizBuySell, and Zillow, applies regional market intelligence and prediction market signals, detects hidden gems, and produces ranked HOT/WARM/WATCH deals with outreach scripts. Use when user says "run hunt", "find deals", "scan for opportunities", "/hunt", "run the pipeline", "find me deals in Nevada County", "what deals are available today", "run all zones", or "start the deal pipeline". allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Agent, WebSearch, WebFetch metadata: author: Deal Intelligence version: 2.0.0 category: workflow-automation tags: [deal-finding, motivated-seller, acquisition-intelligence, any-asset-class]

Deal Intelligence Pipeline

CRITICAL

This system finds deals across ANY asset class — real estate, businesses, auctions, notes, land. It is NOT limited to commercial real estate. Never narrow scope.

Instructions

Step 1: Load Geo Targets

Read data/targets/geos.json. Extract active zones: NCA, CO, USA.

Step 2: Parallel Data Collection

Spawn one data-collector agent per zone + source combination. Run all simultaneously — do not wait for one before starting another.

For each zone, scan all sources in references/sources.md:

  • CREXI (all asset types, not just commercial)
  • Auction.com + Hubzu (foreclosure, REO, tax default)
  • BizBuySell (businesses with real estate)
  • Zillow (residential distress, 90+ DOM)
  • County tax auction listings (WebSearch)

Save raw output: data/raw/YYYY-MM-DD_[zone]_[source].json

Rate limit: 3 seconds minimum between page requests. Stop immediately on 403/429.

Step 3: Regional Intelligence

Spawn market-analyst agent per zone.

For each zone, pull:

  • Population trend (Census QuickFacts)
  • Employment data (BLS by county)
  • News: employers, infrastructure, development (WebSearch "[city] 2026 economic development")
  • Building permit activity

Attach regionalScore (0-100) + 3-sentence summary to each deal.

Step 4: Prediction Market Signals

Spawn prediction-market-intel agent.

Fetch current data from Kalshi and Polymarket:

  • Fed rate decision probabilities
  • Recession odds
  • California housing index (for NCA zone)
  • Colorado economic indicators

Output marketSignal object + timing recommendation per deal. See references/market-signals.md for interpretation guide.

Step 5: Hidden Gem Detection + Scoring

For each deal, compute:

finalScore = dealScore(0.4) + regionalScore(0.35) + marketBoost(0.25) + bonuses

Flag isHiddenGem: true if 2+ criteria met from references/scoring-rubric.md. Hidden gem bonus: +15 points.

Tier classification:

  • HOT (75+): Call today. Pre-write LOI.
  • WARM (55-74): Deep research. Prepare offer.
  • WATCH (35-54): Monitor.
  • PASS (under 35): Archive.

Step 6: Deal Packages

For every HOT and WARM deal, generate:

  1. One-page deal summary (asset, numbers, why it's interesting)
  2. Outreach script — phone + email, seller-specific tone
  3. Offer structure (owner finance terms)
  4. Risk flags and verification checklist

Step 7: Save Report

Save to: data/exports/YYYY-MM-DD_deal-intelligence-report.md

Report must include:

  • Pipeline summary (date, sources, total found, tiers)
  • Zone NCA: Grass Valley / Nevada City results
  • Zone CO: Colorado results
  • Zone USA: Top 3 nationwide (must be different asset classes)
  • Hidden gems section
  • Prediction market context + timing recommendation

Examples

Example 1: Full pipeline

User says: "run the pipeline" Actions: Read geos.json → parallel scan all zones → regional intel → prediction markets → score → packages → save report Result: Report at data/exports/ with HOT/WARM/WATCH deals

Example 2: Single zone

User says: "/hunt NCA" Actions: Same pipeline, NCA zone only Result: Nevada County CA deals report

Example 3: Deal type filter

User says: "/hunt auction" Actions: Pipeline filtered to auction sources only, all zones Result: Auction-only deals report

Error Handling

  • 403/429 from any source: Stop that source, log to data/raw/errors.log, continue others
  • Empty results from a zone: Log warning, continue pipeline
  • Missing geos.json: Create from references/default-geos.md and continue
  • Agent timeout: Log, mark stage incomplete, continue to next stage

Common Issues

If pipeline produces no HOT deals:

  • Check if DOM threshold is too strict (lower from 180 to 90 days)
  • Verify prediction market signals aren't suppressing scores
  • Run /hunt hidden-gems to re-score with relaxed criteria

If data collection fails:

  • Check rate limiting — add 5-second delays
  • Try WebSearch as fallback for any blocked source

References

  • references/sources.md — Full data source list with URLs
  • references/scoring-rubric.md — Complete 6-category scoring breakdown
  • references/market-signals.md — Prediction market interpretation guide
  • references/geo-thesis.md — Zone investment thesis and context
Install via CLI
npx skills add https://github.com/JeromyJSmith/deal-intelligence --skill deal-pipeline
Repository Details
star Stars 0
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator
JeromyJSmith
JeromyJSmith Explore all skills →