Skip to content
Back to search

appeal-draft

Live: https://appeal-draft.vercel.app

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 20 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/2016judea/appeal-draft with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/2016judea/appeal-draft'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/2016judea--appeal-draft/badge.svg)](/listings/2016judea--appeal-draft)

HTML

<a href="/listings/2016judea--appeal-draft"><img src="/api/listings/2016judea--appeal-draft/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 115 lines
  1. # Appeal Draft
  2. **Live:** https://appeal-draft.vercel.app
  3. Type a Twin Cities address. You get the county's value for the home (blue), what recorded
  4. sales of homes like it say it is worth (green), your ZIP's sales per square foot (sand), and
  5. one comparable sale. For $90 you get every comparable sale behind the green number and a
  6. one-page appeal filled in with your county's form fields, steps and dates, ready to print
  7. or save as a PDF.
  8. The next appeal window opens with the **March 2027** valuation notices: local boards and Open
  9. Book meetings run April 1 to May 31, 2027, county boards in June 2027. A Tax Court petition on
  10. the 2026 value is open until April 30, 2027. Sources and read dates: `data/deadlines.json`.
  11. Covers houses and townhouses in Ramsey, Anoka, Dakota, Scott and Washington counties and the
  12. City of Minneapolis (618,892 homes, 48,604 recorded sales since July 2024). Not covered:
  13. Hennepin outside Minneapolis, Carver, condos. See `DATA_MODEL.md` for every field and what
  14. reality could not fill.
  15. ## Ask an AI: the connector
  16. **MCP:** `https://appeal-draft.vercel.app/mcp` (Streamable HTTP, no auth). For the homeowner who asks
  17. Claude or ChatGPT "is my property tax assessment too high, should I appeal?". Three read-only tools:
  18. - `check_property_assessment`: address in; the county's value, what recorded sales of similar homes
  19. say, the gap, a verdict, the yearly tax a cut would save, the 2027 dates, and a link that opens this
  20. page filled in for that home (`/?pk=…&zip=…&sh=…&src=mcp`, plus `&v=` when the owner passes the
  21. value from a newer notice). No ZIP? The Census geocoder finds it; out-of-area addresses get a
  22. plain "not covered", never an estimate.
  23. - `property_tax_appeal_dates`: the 2027 cycle per county, every date with its official source
  24. (`data/deadlines.json` → `cycle_2027`, read 2026-11-19).
  25. - `assessment_ratio_by_area`: the fresh number below, for a ZIP, city or county.
  26. Every number comes from the same shards and files the page reads; a model is never the source of a
  27. number. Nothing is sold in the chat: the comp set and the letter stay on the site. Listing copy and
  28. what needs a click: `SUBMISSION.md`. What it records: `ANALYTICS.md`.
  29. ## Assessment ratio: the fresh number
  30. **Live:** https://appeal-draft.vercel.app/assessment-ratio/
  31. County value ÷ sale price for 17,606 recorded sales, each tested against the value it did not go
  32. into (sales from the October before to the June after the January 2 valuation date). 31% of homes
  33. that sold were valued above their price; the typical home at 0.958. By ZIP the share runs from 14%
  34. (55449, Blaine) to 47% (55130, Saint Paul) among ZIPs with 80+ sales, and the cheapest quarter of
  35. Ramsey homes is valued above its price 49% of the time against 29% for the dearest. Every county
  36. passes the IAAO median and COD tests; Ramsey (PRB −0.051) and Washington (−0.058) sit just past the
  37. PRB line. `scripts/build_ratio.py` → `data/ratio.json`; the page, `llms.txt`, `llms-full.txt` and
  38. the sitemap come from `scripts/build_connector.py`.
  39. ## Expectation
  40. Written 2026-11-19, before the connector is in any directory.
  41. - **Base rates.** heistclips' connector links brought 151 people in 4 days (11/16–11/19), 54% of
  42. whom made a clip, because clippers already ask an AI that exact question. The control, the Brick &
  43. Mortar business connector on the same channel, got zero human calls in 7 days. Directory review
  44. has taken weeks (Claude) and three rounds (ChatGPT).
  45. - **Expected by the read date:** between now and March 2027, about zero human calls (the control's
  46. rate; nobody has a notice in hand). In the two weeks after notices mail: 10 to 40 people landing
  47. from `src=mcp` links and 0 or 1 paid, if at least one directory has listed it by then.
  48. - **Would surprise us, upward:** 100+ `src=mcp` people by the read date, or any paid letter from a
  49. `src=mcp` visit before March.
  50. - **Would surprise us, downward:** zero non-test `mcp_tool_call` events between 2027-03-15 and the
  51. read date while a directory lists it. That says homeowners do not take this question to an
  52. assistant, even on the date.
  53. - **Read on 2027-04-01** (two weeks after mid-March notices): `mcp_tool_call` by client and verdict,
  54. people with `src=mcp` landings and their `checkout_open` / `paid`, and chatgpt.com / bing
  55. referrers on `/assessment-ratio/`. Queries: `ANALYTICS.md`.
  56. ## How good is the green number
  57. Every home that sold itself since January 2025 (30,224 of them; a home is never its own comp)
  58. is a test: the median sale landed at 1.00× our number, and 53% sold within 10% of it. So a gap
  59. under 10% is a phone call to the assessor, not a fight. The comp rule ignores lot size and
  60. water frontage, so a lake lot or acreage reads too low; the page says which sales it used.
  61. Recomputed on every build (`site/public/data/meta.json` → `backtest`).
  62. ## Price
  63. **$90, once, per home. A hypothesis.** Among the 86,790 homes where the county sits 10% or more
  64. above the comps and a tax bill is on the roll, the median yearly saving if the value fell to
  65. the comp number is $954 (tax × the cut, assuming tax moves with value; it moves less). A tenth
  66. of one year's saving is about $95; $90 is the round number under it. Measured 2026-10-19.
  67. Counter-signal: AppealDesk sells a DIY evidence packet for $49 flat (their own comparison page,
  68. read 2026-10-19). If round 1 shows checkout opens and no pays, $49 is the next test.
  69. ## Distribution
  70. Round 1 is drafted, not sent: 50 recipients in `outreach/round1/` (gitignored), each with a
  71. `/for/<token>/` page that opens on a home of their own — 40 listing agents (their live
  72. listing), 10 small landlords (a house their LLC owns), and 5 MN appeal services on hold (no home
  73. of their own to open on). Subject `bricks / <company>`.
  74. A/B, the same single variable on all seven shelf products so results pool: **A** states the
  75. price in the email, **B** leaves it to the page. Alternated within each segment; the variant
  76. rides in `/data/for/<token>.json` and fires with `preview_open`, never in the URL.
  77. Hypothesis: stating the price in the email lowers clicks but raises paid-per-click. Readout on
  78. day 3: visits, checkout_open, paid and replies, by variant, by segment and pooled.
  79. ## Run it
  80. make build # closed_system.duckdb (read-only) -> site/public/data + private/bundle, the ratio study, the pages
  81. make page # just the pages (index, /assessment-ratio/, /privacy/, llms*.txt, sitemap, api/_mcp_data.json)
  82. make test # pytest + node handler tests (paywall and /mcp), logs/*.log
  83. make serve # localhost:4321 with the API handlers and /mcp
  84. set -a; . ~/Desktop/workspace/bricks/.env; set +a; make deploy # then pings IndexNow
  85. python3 scripts/build_chatgpt_plugin.py # chatgpt-plugin.zip for the ChatGPT plugin portal
  86. Only `site/public/` is served statically. The paid comp sets ride up as `api/_full/`, bundled
  87. with `/api/data` and never under a public URL. Webhook goes through `~/Desktop/workspace/stripe-router`.
  88. `scripts/fulfil.py` reprints a lost password or mints a comp one.
  89. MIT licence for the code. The data is public record from the county assessors and recorded
  90. sales; the owner name is never printed, the homeowner types it on the letter.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.