SS—03

THE LIVE STACK
COMPOSER 2.5 · OCT 2026

… of … hosts answering

The Lab

MEDICINE / MILITARY / MACHINESINDEPENDENT WORK, 2026 ↘

SEAN SOOCH.

Three disciplines.
One restless curiosity.

Medical student. Military officer. Builder.
A collection of tools, research, and experiments made with frontier AI, in the hours between everything else.

Explore the work ↙
THE CURRENT EDITION03CURSOR COMPOSER 2.5 / LIVE STACK ↗︎
And everything
in between. ↗︎

01THE WORK

Built to be used.

14 selected systems and the music. Open one and put it to work.

01LiveMedicine

Step 2 CK QbankRebuilt

Public study site over seven Anki-derived banks: 12,972 board-style questions and 12,357 guide facts across eight subjects, a UWorld-style test engine with tutor and exam modes, per-system performance, and three drill tools. Built by a chunk, author, independent-review, adversarial-triage pipeline that caught and corrected about 100 factual errors in the commercial source decks.

12,972 questions8 subjects2-3 review passesOpen → 02Live · ~30s cold startMedicine · QI

Patient Handoff Tracker

Full-stack I-PASS/SBAR handoff system built as an emergency medicine QI project: named-receiver read-back that only the receiver can acknowledge and verify, action items, Joint Commission compliance metrics with CSV export, and HIPAA-conscious audit logging over synthetic patients. Flask and SQLite on a free tier, so the first load wakes the server.

22 API endpoints8 DB tablesFlaskOpen → 03LiveMedicine · Cardiology

Read This

Pattern-interpretation trainer: 28 procedurally generated rhythm strips at 25 mm/s with calipers and a live monitor, a fixed 8-step read of 14 twelve-lead cases with MI localization by territory, and an outside-in chest film read over synthesized radiographs and real films. Every case sourced.

28 rhythms8 read steps100% sourcedOpen → 04Live · best on desktopMedicine × Game

MICRO TD

Tower defense where the body is the map: 16 guideline cases as levels, 34 real drugs as towers against 25 organisms, damage set by coverage and site penetration, resistance bred by antibiotic pressure, and organ toxicity, plus a one-patient Ward Rounds mode. TypeScript and Canvas, every fact sourced, 104 tests.

16 guideline cases34 drugs25 organismsPlay → 05LiveMilitary

Warrior Ready

All-branch PT scorecard at 2026 standards: six services scored from their official charts with the source cited, per-service body composition screens, run pacer with lap splits, ruck planner, and score history. The build refuses to ship unless 72 spot checks read off the printed charts pass.

6 branches72 chart spot checks0 depsOpen → 06LiveMachines · AI Safety

AGI Capability Tracker

Frontier capability ledger across 10 sectors, 57 models and dozens of benchmarks: a weighted composite score with its method published, verified ARC-AGI cells, a capability-milestones register, a capability-against-control safety ledger, and a gated-tier toggle for what the labs hold back. Updated September 2026.

57 models10 sectors5 milestonesOpen → 07LiveMachines × Medicine

Simulators

Four sims, one umbrella: the Alignment Simulator (run a frontier lab, 12 decisions, 6 endings, built from the Risk Stack paper), Code Blue (spacebar CPR with live rate and depth feedback, AHA 2020), Half-Life Tracker (caffeine, hydration, and BAC pharmacokinetics), and Pocket Telescope (live night-sky ID and star quizzes).

4 sims24 decisions11 endingsRun → 08LiveSports · Fantasy Football

Edge Draft Terminal

Draft intelligence terminal for a 12-team half-PPR league: an 862-player board valued on a weighted seven-source composite and read against the Sleeper room's market price, weekly projections priced as points above replacement, Monte Carlo pick-path simulation in a worker, an in-season wire of injuries and market movers, and read-only replay of a live Sleeper draft room. Static PWA with CI-gated deploys and a test suite that also runs against production.

862 players7 sources21,761 lines of testsDraft → 09LiveSports · Hoops

NCAAT Bracket

Archive of the finished 2026 men's tournament: all 67 games from a frozen ESPN snapshot, every first-edition model graded against the results, and an interactive bracket that scores your own picks. A sourced fact-check corrected 531 first-edition claims, all listed on the page.

67 games4 models graded531 correctionsOpen → 10LiveSports · Golf

Masters 2026 Hub

Archive of the 2026 Masters as a hand-operated leaderboard: the final board for all 91 players with hole-by-hole cards from a frozen ESPN snapshot, an 18-hole yardage-book course guide, a searchable field, the champions' roll, and a Remotion film rendered from code. 137 corrections to the first edition are listed.

91 players18 holes600 framesOpen → 11LiveSports · Detroit

Detroit Sports Hub

Hub board for the Lions, Tigers, Pistons and Red Wings plus three trackers: a new Lions app (live down and distance on a drawn field, drive charts, win probability), Tigers with pitch-by-pitch tracking, Statcast and a 30-prospect farm system, and the Pistons' 14-68 to East #1 seed run with injury watch and a 2004 tribute.

3 trackers29 tabs0 depsOpen → 12LiveHumanities · Literature & Philosophy

Great Books

Continuous-zoom library of 46 works across existentialism, political order, liberty, machines, knowledge, and reality.

46 books5 shelves4 depthsRead → 13LiveMedicine · Education

AdmissionsAcademy

Founded by Paul Clancy to expand access to medical education: a free Pre-Med Essentials course, MCAT strategy, and application guidance for students from any background. Sean serves as Chief Technology Officer.

Free premed course1-on-1 guidanceCTO Sean's roleVisit → 14LiveMusic

Music

The Claudia Bench: anabology's Claudia films, kept as the bar. Measured against it, our own: LAST LIGHT, a grunge countdown to the Sun's last light; The Glass, the Claudia character singing track 4 of the Three Roads EP; and Til It's Over, the newest song.

3 bench films3 songs of ours1 paper sungListen → 15LiveMachines · The Internet

Monitoring the SituationNew

One record of what every corner of the internet is passing around. Grok reads X each hour across twenty desks; Claude confirms every post with X, writes a line on what it is, and researches the larger stories, setting out how each world describes them and what is on the record.

20 desks2 AI agents2 checks per postOpen →

NEWTHE LAB

Medicine, one step at a time.

visual lessons in plain language · educational, not for clinical use

A course in core medicine, taught one picture at a time. Each lesson takes a single idea, starting with how the heart fills and empties, builds it step by step in plain language with animated visuals, then shows how common drug classes change the picture. New lessons are added in order. Educational only, not for clinical use.

LAll lessonsThe course page: every lesson in order, each one a single idea built up step by step.Course · MedicineEnter →

02THE WRITING

Working theories.

3 papers · 2 reading rooms · 19 essays

Humanities · Independent researchTable 4-1 · April 2026

The Frontier AI Risk Stack

Frontier AI risk is best understood not as a single category but as a stack of six interacting layers. The most serious dangers arise from the way failures compound across layers: a structured analytical tool for policymakers, developers, and civil society.

Read on AGI TrackerTechnical PDFPlain-language PDF
  1. 01Model Riskdeception, unsafe knowledge, concealment
  2. 02System Riskorchestration, tool use, parallel execution
  3. 03Deployment Riskreal permissions, production access
  4. 04Access Riskconcentration, cognitive inequality
  5. 05Governance Riskweak audits, regulatory capture
  6. 06Civilizational Riskcompounding failures across all layers

Humanities · Independent researchTable 4-2 · June 2026

Three Roads From Here

The three positions on frontier AI (safety through development, acceleration, pause) characterized on their strongest terms, with existential and suffering risk treated as separate axes that do not always move together. The central frame: the camps are reading three different payoff matrices for the same game.

Academic PDFSimple Mode PDFGame Theory Mode PDF
  1. 01Steerbuild inside guardrails and tripwires
  2. 02Speedthe race cannot be stopped, only won
  3. 03Stoppause until alignment is solved
  4. 04X-risklosing the future entirely
  5. 05S-riska future that continues, and hurts
  6. 06The gameone matrix, three readings, edit the payoffs

Humanities · Independent researchTable 4-3 · September 2026

Two Stories About the Beginning

The Big Bang and a generated origin are the same kind of story: a starting condition plus a rule, with the evidence only ever reaching the rule. Four objections (tidiness, cheating, motive, proving too much) are taken seriously; two of them force the second story to change, and it survives as a testable rival rather than an evasion.

Read the paperTable 4-3, side by sidePart II, stated in full
  1. 01Start + ruleevery origin story, and the evidence reaches only the rule
  2. 02Tidinessthe first story already carries the extra layer
  3. 03Cheatingno shortcuts; the saving is that the rule is short
  4. 04Motivedropped; the process only has to exist and run
  5. 05Last Thursdaya rival commits to something, an evasion to nothing
  6. 06The testsfinite grain, finite information, more room outside
Annotated readingFig. 4-1

Reading Room

An annotated card catalogue of 64 AI papers, X threads and blog posts collected through April 2026, each with a technical summary, a plain-language version and a note on why it matters.

papers · tweets · blogsEnter →19 essays · 6 sectionsFig. 4-2

Philosophy Room

One thesis: every great question converges on alignment, whether we can build minds that want what we want. Cosmology, consciousness, AI, identity, theology and game theory, each question read twice: as the classical problem and through the AI lens.

~30 min readEnter →

03THE FILMS

Press play.

rendered from code, captured from live sites · sound on

Plate 1. Launch film. Vol. 03 in motion: the cover, the topology answering, the roll call, the record, the index.1:20 · sound on

04THE PERSON

In the spare hours.

1-1.Purpose

This manual documents five months of learning and building in spare hours. These projects were made possible by tools that lowered the technical barrier between an idea and working software. Others will bring different experience and build better things.

1-2.Scope

Everything listed is deployed and public. Chapter 2 does not take that on faith: when this page loads, your browser pings every host in Figure 1-1 and records what came back. Chapter 3 lists the systems by domain, Chapter 4 collects the written doctrine, Chapter 5 the smaller equipment. Appendix A is the change record for the manual itself.

1-3.Method

Every system was specified by the operator and written with Claude, then verified the way a clinician verifies a lab: against a source, by a second reader, and against the live deployment rather than the local copy. Medical content carries its citations. Scoring tables cite the governing publication. Where a source deck was wrong, the question was corrected and the correction logged.

NOTE

Study tools on this site are for exam preparation and are not medical advice. Clinical calculators and field references are teaching aids; confirm with the governing guideline before use on a patient.

05THE EXPERIMENTS

And a few detours.

smaller experiments, same operator, same period

5-1 ForgePrompt-to-HTML site builder on the Anthropic streaming API, clone mode, component injector, live theme controls.Claude API · VisionOpen →5-2 Hospital & Field ToolkitClinical calculators, antibiotic coverage matrix, differentials, lab reference with the CBC + CMP decoder, and TCCC MARCH: five tools, one hub.Medicine · TCCCOpen →5-3 Space ExplorerLive dashboard in eight sections over twelve public feeds, seven of them NASA's: APOD, the ISS over a map with its ground track, launch countdowns, EPIC Earth imagery and events, the Sun and its flares, near-Earth approaches, deep space, news.NASA · Live APIsExplore →5-4 Force MultipliersThree briefing cards on the tools behind the work: graph versus loop engineering for AI agents, the reader in Obsidian, and the scheduler in Anki. Each ends with three check questions.Methods · AI agentsOpen →

06THE EDITIONS

Same person. New perspective.

one redesign per model class · the site is the benchmark

This site is rebuilt from the ground up at every new Anthropic model class, as a personal benchmark of frontier capability. Each edition is designed end to end by the model it is named for, then retired, not deleted, when its successor arrives. The archive is the measurement: put two volumes side by side and you can see, in typography and judgment rather than benchmark scores, how far the frontier moved. Vol. 01 wore the newsroom, Vol. 02 wore the clinic, and Vol. 03 wears the uniform.

Table A-1. Editions of this manual.

ChangeEditionDesigned byIssuedStatus
Vol. 01The Magazine: dark editorial, sixteen sections, terminal metricsClaude Opus 4.7April 2026Retired with honorsView archive →
Vol. 02Triad × Vitals: three worlds, one chart, reactive inkClaude Fable 5June 2026Retired with honorsView archive →
Vol. 03Four views: the Field Manual, the Personal Index, the Commentary and the Live Stack, switchable in placeOpenAI Astra + Claude Fable 5.1 + Claude Opus 5.5 + Cursor Composer 2.5September 2026CurrentYou are here
Vol. 04To be designedThe next class-Awaiting release

Same world. Four views.In the order they joined. The switch at the foot of every page keeps your place.

01Claude Fable 5.1 · Vol. 03The Field ManualNumbered chapters, live readiness, a printed index.02OpenAI Astra · Vol. 03The Personal IndexA white exhibition catalogue: the name set huge, numbered editorial rows.03Claude Opus 5.5 · Vol. 03The Commentary Frontier leadA glossed edition: the work in the centre, commentary in the margins, one rubric red.
04Cursor Composer 2.5 · Vol. 03The Live Stack NewAn operations deck: dark telemetry, green reachability lights, monospace labels.You are here
NOTE

What changed in Change 3, beyond the design: the Step 2 site was rebuilt as a public question bank (12,972 items, up from 2,245), the lab reference gained a clickable CBC + CMP decoder, the AGI Tracker gained a milestones timeline, Edge gained its in-season wire, and the line count was recounted from source with generated data excluded (150,906, up from 65,538). Two things this edition adds that no earlier volume had: live reachability measured from the reader's browser, and an index.

07THE RECORD

Open for inspection.

counts recounted 14 Sep 2026 · reachability measured live on every page load

Source figures, live status & full directory +

Table 2-1. Portfolio readiness. Static figures are recounted at every edition; live figures come from Table 2-2.

MeasureResultStandard
Lines of code150,906unique source files, generated data excluded; 21,761 of these are one project's test suite
Source files247html, js, css, python, typescript
Board-style questions12,9727 question banks, 8 subjects, every item independently reviewed, most a second time
Players & teams tracked1,106sports intel across 12 data sources
Systems reachable… of 17measured from your browser, now
Median round trip…milliseconds, edge-cached static hosts; the one server-rendered app is excluded while it wakes
Months since first deploy5April to September 2026, evenings and weekends

Table 2-2. System status. Same probe as Figure 1-1, in list form.

SystemHostDomainStatusRound trip
CAUTION

The probe is an opaque cross-origin fetch, so it can only report that a host answered, not that the answer was the right page. Cloudflare Pages serves a soft 200 for unknown paths. Treat green as "the lights are on" and click through for anything that matters.

All twenty-one systems

Checking reachability…

A—Z

The complete index.

every figure, table, and film on this page, alphabetically · press / to search it