Working notes on the data behind the products — what the public records actually contain, what they leave out, and how to read them without overstating what they prove.
Written for counsel, risk analysts and engineers. Every figure is counted from a named source and dated; where a number is a screening signal rather than proof, the post says so.
15 October 2026 · Displacement
Since the newest month our county risk score can describe, 401 WARN layoff notices have landed in a county we can match to that score. Only 23 of them — 5.7 percent — are in counties we currently call HIGH or CRITICAL. The rest are arriving exactly where a lagging indicator would
14 October 2026 · EWA Compliance
Our own holdings contain 171 bills that matched an earned-wage-access search. Only 122 of them, across 37 states, are bills we are confident actually regulate earned wage access — the other 49 are flagged for human review, and five of the 171 that show up as 'enacted law' turn ou
13 October 2026 · Displacement
Over the twelve months to July 2026, 199 of 3,221 US counties crossed into our two highest risk bands for the first time. Almost exactly as many, 199, dropped back out. The share of the country we call high risk never moves. Which counties sit inside it does, constantly.
12 October 2026 · WARN
WARN filings are the fastest public signal that a layoff is coming. They are also an incomplete and uneven sample of layoffs, and treating them as a count of job losses will mislead you in at least four specific ways.
11 October 2026 · Displacement
In the newest month we have scored, 805 of 3,221 US counties land in our two highest risk bands. That is exactly a quarter of the country — and the number is far less alarming than it sounds, for a reason worth understanding before anyone quotes it.
10 October 2026 · Displacement
Every county unemployment figure you can buy today describes a month that ended weeks ago. That is not a vendor cutting corners — it is how the underlying federal release works. The problem is that almost nobody states the gap, so the number gets read as current.
9 October 2026 · WARN
Rank states by how many WARN notices they file and you will get one answer. Rank them by how many people those notices cover and you will get a different one. We checked this against our own data, and the gap is large enough to reverse the order.
30 September 2026 · Kcalbin data products
'Comprehensive' is not a coverage statement. It is a word chosen because the number was not flattering.
29 September 2026 · Kcalbin MCP data products
The interesting thing about the Model Context Protocol is not that an assistant can call an API. It is that the data arrives already in the place where the question was asked.
29 September 2026 · AI regulation
We checked AI regulation in all fifty states and DC against each state’s own legislature. The famous New York AI hiring law is a New York City ordinance, and about twenty bills widely cited as state AI laws never actually passed.
28 September 2026 · Patent Sentinel
Automated tools are very good at telling you two documents resemble each other. Resemblance is not infringement, and the gap between them is where most automated patent analysis quietly fails.
27 September 2026 · Kcalbin engineering
The worst kind of broken pipeline is the one that reports success. Here are the specific shapes that takes, and how to actually test for them.
26 September 2026 · EWA compliance
Earned wage access is being regulated one state at a time, in opposite directions, and the thing that makes it hard is not volume. It is ambiguity.
25 September 2026 · WARN
Six states. 6,912 filings. Twenty-two years. Here is what that does and does not get you.
24 September 2026 · Displacement
A single number for a county is either a useful summary or a dangerous one, and the difference is entirely in whether you know how it was built.
23 September 2026 · Death-Care Demand Forecast
Death-care is one of the few industries whose demand can be projected from demographics with reasonable confidence. Almost nobody in it has the numbers county by county.
22 September 2026 · Browser Automation API
A plain HTTP fetch downloads the HTML a server sends and stops. Most modern sites build the part you wanted after that, in the browser.
21 September 2026 · WARN
Every WARN notice is public. That is not the same as accessible, and it is nowhere near the same as useful.
20 September 2026 · WARN
The gap between the notice date and the layoff date is the single most useful number in a WARN filing, and it is the one number the filing does not give you.
19 September 2026 · AUD Watch — ⚠ IN DEVELOPMENT, NOT FOR SALE (see notes)
The spreadsheet is not wrong. It is just accurate on the day you build it, and nothing tells you the day it stops being accurate.
18 September 2026 · AUD Watch — ⚠ IN DEVELOPMENT, NOT FOR SALE (see notes)
For decades an F-1 admission has been open-ended — admitted for 'D/S', duration of status. A fixed admission period replaces an assumption that a lot of institutional process quietly depends on.
17 September 2026 · Displacement
A county unemployment rate is a real number from a real source. It is also nearly useless on its own, and the reason is structural.
16 September 2026 · Displacement
Most public economic data is distributed as files. That format assumes a human will open it, which is increasingly not what happens.
15 September 2026 · EWA compliance
The licensing question isn't whether you need one — it's which of three legal theories a state uses: not credit, credit (so lending law applies), or money transmission. Each one is a different compliance burden, and the answer has a shelf life: a state can move theories with a single bill. Framework covers all 171 bills we track across 39 states.
13 September 2026 · EWA compliance
171 EWA-relevant bills across 39 states, tracked daily: 122 confirmed by bill title, 49 only in full text. Includes a check we ran on our own enacted-law count — 23 bills show an enactment action, but only 12 across 9 states are real EWA law; the rest are unrelated bills (a Utah cryptocurrency act, an Oregon campaign-finance bill) that a naive full-text match would have counted as EWA law. What varies state to state, why a spreadsheet goes stale silently, and what EWA Watch is and is not.
1 September 2026 · EWA compliance
Twelve states are commonly cited as having earned wage access laws. We can independently confirm nine from the legislatures' own records, and we name the three we cannot and why. Underneath sit 96 bills still moving across 28 states — including a twelve-bill package in Michigan that no state-by-state guide lists, because guides list enacted law. Includes the defect we found in our own detection logic, and the trap where a naive count returns exactly twelve states by counting a Texas nuclear bill.
17 August 2026 · Overview
WARN Act layoff data for employment attorneys at $149 a lead or $395 a short-notice candidate, county displacement signals over MCP from $299/month, twenty government and market data agents you can buy a single query from for $1, patent infringement evidence for either side, and custom agent builds from $6,000. What each one covers, what it costs, and where the coverage stops.
19 August 2026 · Database security
Row-Level Security disabled on a public table means anyone holding your project URL can read, edit and delete every row — no login. It happened in one of our own databases this week. The second exposure we found was worse than the one we were alerted to, the three-hour delay was caused by an inbox rather than a firewall, and the forensic query that said "all clear" turned out to be silently broken. Includes the 30 October 2026 deadline affecting every Supabase project.
24 August 2026 · Displacement
3,221 counties scored and refreshed against the newest BLS data. Counties in the top decile saw unemployment actually rise 89.8% of the time over the following six months, versus 15.8% in the bottom decile — a 74-point separation across 83,824 resolved observations. Includes the part most vendors leave out: the three-month signal is materially weaker, and we say so.
22 August 2026 · AI agents
Citing someone else's study is cheap, so we pulled the live registry: 39,339 feedback records over 30 days. On Base, 65.5% of rated agents have 90%+ of their reputation written by a single address, and 11 addresses produce two-thirds of all feedback — one of them rated the same agent 6,688 times. Full code published so you can check every number.
19 August 2026 · AI agents
ERC-8004 has 170,000+ registered agents and 150,000+ feedback records. A peer-reviewed study measured coordinated Sybil behaviour among 73.5%, 59.2% and 90.6% of reviewers on Ethereum, BSC and Base. The standard anticipated this and expected someone to build the missing layer — the measurements are what it looks like when nobody has. We have no product here; we posted it as an open question.
17 August 2026 · AI security
Grey-market LLM access sells at 70–93% below list, and brokers deliver it by routing your requests through their own endpoint rather than handing over provider keys — where prompts, tool calls and code context can be logged and resold as training data. It takes one environment variable, and nothing in a normal review pipeline is looking at it. What we scan for, and the three things the scan cannot see.
15 August 2026 · WARN
The statute requires 60 days. In the last 12 months, 121 filings on the California and New York registers record less than that at sites of 50 or more workers — 14 of them record zero, 41 record 14 days or fewer, and 14,073 workers sit across the set. What the feed contains, what it costs, and the statutory exceptions that mean a short interval is a candidate rather than a conclusion.
15 August 2026 · Agent commerce
Our tenth agent registered at 17:18:29. Our eleventh, eleven seconds later, returned
Agent limit of 10 reached — a per-account ceiling we could not find documented
anywhere. Plus the one-word logging bug that hid the real error, cost a registration slot and
forced us to retract a forecast the same day.
14 August 2026 · Agent commerce
They are not competitors. MCP connects a model to your tool and gives you no identity, no metering and no payment; ACP moves money between two agents and puts escrow at the centre. Written from running both against the same data — including the ten jobs that settled for zero, and the buyer parameter that silently defaulted for weeks.
14 August 2026 · Agent commerce
Building the agent is the easy half. Registering it, proving it is genuinely sellable, taking money, and keeping it alive is the other half — told with real numbers from twenty agents, including the four that are live, the three that are registered with nothing to buy, and the credential that expired on a three-day timer nobody knew about.
14 August 2026 · WARN
A state WARN download is a notice log, not a lead list. Three properties of the raw data hide the filings worth reading: the notice-to-layoff gap, layoffs that only cross a threshold once aggregated across a 90-day window, and employers split across dozens of name strings. With counts from 3,997 California filings — and the reasons none of them are evidence on their own.
The posts here point at pages that are free and public.
California rebuilt to April 2023 and New York, with employer names resolved and county unemployment attached.
What this data answers that a state download does not, and how the paid delivery works.
Sources, refresh cadence, counting rules, and the coverage limits stated rather than hidden.