🔍 Search
Type an address, a parcel number (with or without dashes) or an owner name. Misspellings and extra words are fine; add a county or state to search every parcel there, not just the lots for sale.
Investors on here who claimed the portfolios they own. The one holding the lot you want is often already on this list.
🏷️ Buy without an auction
Some counties and states sell property over the counter, by sealed bid, or as surplus after an auction fails. These programs are state- and county-specific — find yours here.
Loading deals…
AI job queue
Analysis & county-assessment jobs run one at a time. Paused by default — resume to process the queue. This pause only covers jobs requested here; scheduled crawlers (e.g. iltaxsale's daily extraction) call the AI backend directly on their own timer and keep running regardless.
AI orders & call history
Every AI call/order — backend, model, tokens, cost, status.
AI engine
This is the AI that reads each uploaded file and evaluates your parcels — it maps the CSV columns on import and produces every parcel's 0–100 score, tags, and rationale. Choose where it runs and which model it uses; changes apply immediately, no restart.
Usage analytics
Claude analysis report
Run the AI evaluation and geocode enrichment for a whole import, and read the resulting summary. Analysis runs through the AI job queue.
Teams / workspaces
Buyer portfolio claims
"that owner_key is me" — approve only with evidenceTeam billing & credits
$100/mo + $75/seat · auto-paid from creditUsers
Free parcel-data sources
Query a county/state ArcGIS parcel service by APN to fill geometry, address, owner, etc. — no paid key. Give the layer query URL (…/FeatureServer/0 or …/MapServer/0) and the APN field name.
Auction platform accounts
Gated auction platforms need a real login (and, for crawl4me headless crawling, the operator's browser-extension consent). Track which accounts we need vs. have here. No secrets are stored — just status + a note on who holds the login.
Map & imagery API keys
Keys are stored server-side and never exposed publicly. Reload the page after saving for map tabs to update. Leave a field blank to clear its key.
Google enrichment APIs
Geocoding, Places, Routes, Air Quality, Pollen, and Solar all run through the Google Maps Platform key above, but each is its own billed API. Toggle each independently below; a parcel that hits a paused API just shows "No data" for that field until it's resumed. Costs are estimates from public list pricing, not your actual GCP invoice — check GCP Console → Billing → Reports for the real number.
Paid parcel-data providers
Server-side keys for owner / deed / sale-history & (soon) lien lookups by APN. Never sent to the browser. Paid enrichment runs only through the job queue when an admin triggers it — it is not part of the free auto-enrich. Precedence when multiple are set: Regrid → ReportAll → ATTOM.
Parcel API access tokens
Keys we issue for our own parcel API at /api/v2 — point→parcel, address, APN, owner portfolio, area, field query and typeahead, served from our crawled data instead of a paid provider. Tokens are workspace-scoped and stored hashed: the value is shown once, on creation. Browse the API · field schema
crawl4me jobs
Every crawl4me job this app uses — platform crawls, GovEase, and browse-capture jobs that record what an operator browses. Walled sites wait for an operator's browser extension.
+ New job
Crawl monitor
Live view of the crawl4me GovEase job — the county queue, capture progress, and any bad (page-shell) captures that need re-running.
Crawl → AI → parcels
what this shows
Every crawl4me capture is saved before processing, traced through the AI column-mapper to the parcels it produced, then gzipped so the bulk leaves the live disk (kept for audit). Follow any single capture end to end, and watch which counties are healthy.
⚠ Needs attention
what this shows
Only genuinely actionable problems: crawls producing nothing, counties that stopped arriving past their own cadence, sale dates that passed with no ingest, or thin data where we hold real volume. Counties we simply haven't started are tracked as coverage backlog, not alerts.
Import yield
what this shows
Every import with a future (or very recent) auction date, checked against the SAME bar the Deals hunts use: an opening bid or owed balance, a value to compare it to, and a plausible ratio between them. A live import with zero scorable lots is invisible everywhere else — it just never shows up. Listed here only when it's an anomaly — its own state is landing the field this import is missing, so something on that page isn't being read. States with no value or bid source wired yet, and sales that already happened, are counted as backlog underneath instead of alerted forever.
Source health
what this shows
One row per registered source. Alerts = a feed that was working and broke (labelled with the cause + fix, not a bare ✗). Backlog = researched-but-unwired coverage. Population/join % show a source can be reachable yet useless. Drill in for the raw body + lineage.
Needs your input
what this shows
Sources reachable-but-unusable until a human does something code can't: hand over a login, make a policy call, or approve infra spend. Ranked by counties unblocked. Resolved/false-positive rows drop off this list — see research/34-access-blockers.md for the full history.
Domains
what this shows
One row per external domain, collapsed from its subdomains (87 Ohio county sites all roll up to one "sheriffsaleauction.ohio.gov" row). Ranked by counties served, tie-broken by rows ingested — urgent (broken, or sitting behind an open ask) always sorts first.
Crawl volume & health
what this shows
One cell per source per day — thousands of crawls at a glance. Green = all captures produced rows, amber = some skipped, red = failures. Click a cell to inspect exactly those captures.
Captures
County input health
what this shows
How well we are gathering each county: freshness against that county's own sale cadence, crawl reliability, and how complete the data is once it lands.
Maintenance actions
Archiving gzips processed capture bodies off the live disk (kept for audit). Rebuilding health rescores every county's input freshness and completeness.
Deal grading
what this shows
The grade is deterministic math — value ÷ cost measured against the county norm, adjusted for signals and data confidence. No AI. Letters come from percentile cutoffs on the 0–100 score, so a grade means "this good relative to everything we hold", with absolute guards that stop a percentile from flattering a bad deal.
Query cache
what this shows
Browse/sitemap/landing-page counts and the default top-owners view are pre-computed on this schedule instead of recomputed per request. Bumped this after a bulk ingest queued up behind a live county-count scan and stalled login — see the incident note. A stale row (age > 2× the interval) usually means the timer stopped; refresh it manually here.
Collectors
Every connector the daily scan runs. Window is when it runs now; Proposed is when its state actually holds sales (all sale history, widened one month each side), with how many runs a year that would save. Nothing here changes a schedule. Low-confidence rows have little sale history.
Auction-source coverage
Every county's tax-sale source: third-party platforms (GovEase, etc.) vs. counties that publish directly, and how far along we are ingesting each. Filter to find gaps and direct-publishing counties to build next.
Auction platforms
Cross-county auction aggregators — the crawl4me targets. Where sale listings live nationally/regionally, whether they're open or WAF/CAPTCHA-gated (need a headless crawl), and our crawl status. Add sources here as we find them.
Listing verification queue
A listing needs at least one proof-of-purchase document before it can go publicActivity log
What background jobs & automations (imports, the Kansas tax-sale monitor, enrichment, errors) have been doing. Filter by message type.
Banned IPs
Auto-banned by the rate limiter (>3 page/API requests per second from one IP) or the bot User-Agent block. Bans persist across restarts. Unban if this was a false positive (e.g. many users behind one office/VPN IP).
Landing map snapshot
The homepage map reads a cached JSON snapshot (public/demo-map.json) — a dot per active scored lot plus one featured lot. Rebuild it from live data when you want the marketing map to catch up.
“Real lots on the board” strip
The sample cards under the hero. Leave blank to auto-pick the best scored lots, or pin specific parcel IDs (comma-separated) to curate exactly what shows.
Data coverage
Score distribution
By category
Top counties
Hazard risk (FEMA NRI)
Farmland classification
Top AI tags
Risk flags
Bid results
⏰ Upcoming deadlines
Paste this into Google Calendar (Other calendars → From URL), Outlook, or Apple Calendar (File → New Calendar Subscription). Anyone with the link can see your team's tracked sale dates — treat it like a password.
Upcoming auctions
Recent team activity
County assessments
Claude assesses each county in your imports (redemption rules, sale type, market, risks). Your team can comment.
List a parcel for sale
Only parcels marked Won in your Deal Pipeline are eligible. Attach proof of purchase (deed, settlement statement) below — an admin reviews it before the listing goes public on /for-sale.
Your account
Public buyer profile
Claim the portfolios you own and other subscribers can find you — in the markets you buy in, on the parcels you already hold. Private by default; nothing here is visible to the open web.
Change password
Your team
Subscription
Account credit history
API access
Programmatic access to your data (/api/v2)Create a token to query the parcel/deals API from scripts. Treat tokens like passwords — they're shown once.
API reference & quick-start
Spreadsheet feeds
Live CSV links for Google Sheets or ExcelCreate a link, then in Google Sheets use =IMPORTDATA("the link"). Anyone with the link can read that feed, so treat it like a password; revoke it to cut access.
Alert rules
Get notified (🔔) when newly-analyzed parcels match your criteria — e.g. deal score ≥ 70 in a given state. Rules run automatically after each analysis.