Skip to content
FINQUILL://terminal

The Investment
Reasoning Engine

Your conviction: |

Falsifiable theses
Adversarial red-team
A graded record
Situation Monitor · one recorded day

A globe that knows your book.

Most situation maps show you everything and leave the sorting to you. The Situation Monitor knows your watchlist, your theses and your holdings, so it knows which parts of the world are yours: the countries you're exposed to are tinted, and a quake, storm or volcano inside one of them is ringed on the globe and raised in the Monitor's signal feed when it's serious. Everything else stays ambient. Counted from open data, never AI-scored, and viewing costs no credits.

Here the book is Finquill's own — the Board and the names its narratives track — on one recorded day.

Touches this book · Oct 28, 2026

  • VolcanoKilauea Volcano, Hawaii · volcano, open eventUnited States of America · NASA EONET
  • QuakeM5.4 earthquake · 210 km ENE of Nagqu, ChinaChina · USGS
  • QuakeM4.7 earthquake · 81 km ENE of Mazha, ChinaChina · USGS

This book's exposure

China · BABANetherlands · ASMLTaiwan · TSMUnited States of America · 43 names

Open the Monitor

Replaying Oct 28, 2026 · earthquakes USGS · storms NOAA NHC and NASA EONET · volcanoes NASA EONET · fire detections NASA FIRMS · chokepoint transits IMF PortWatch · borders Natural Earth · ships and flights are sample positions, not tracked

Illustrative traffic — sample positions

Replaying Oct 28, 2026 · 58 events · 3 touch this book · drag to spin · hover a pin

  • Strait of Hormuz · 23 transits in 7 days · −11.5% vs the prior week
  • Suez Canal · 289 transits in 7 days · −6.8% vs the prior week
  • Bab el-Mandeb · 202 transits in 7 days · +9.2% vs the prior week
  • Panama Canal · 201 transits in 7 days · +3.6% vs the prior week
  • Taiwan Strait · 1,330 transits in 7 days · −16.4% vs the prior week
  • Strait of Malacca · 1,596 transits in 7 days · +6% vs the prior week
  • Tropical Storm Vance · 35 kt
  • Tropical Cyclone Arnab
  • Tropical Storm Olana
  • Super Typhoon In-fa
  • Tropical Cyclone Agueda
  • Super Typhoon Champi
  • Nevados del Chillan Volcano, Chile · volcano, open event
  • Telica Volcano, Nicaragua · volcano, open event
  • Ivao Group Volcano, Russia · volcano, open event
  • Titan Ridge Volcano, Papua New Guinea · volcano, open event
  • Kikai Volcano, Japan · volcano, open event
  • Bezymianny Volcano, Russia · volcano, open event
  • Langila Volcano, Papua New Guinea · volcano, open event
  • Rincon de la Vieja Volcano, Costa Rica · volcano, open event
  • Masaya Volcano, Nicaragua · volcano, open event
  • Mayon Volcano, Philippines · volcano, open event
  • Sabancaya Volcano, Peru · volcano, open event
  • Ambae Volcano, Vanuatu · volcano, open event
  • Puracé Volcano, Colombia · volcano, open event
  • Kanlaon Volcano, Philippines · volcano, open event
  • Krasheninnikov Volcano, Russia · volcano, open event
  • Popocatepetl Volcano, Mexico · volcano, open event
  • Lewotolok Volcano, Indonesia · volcano, open event
  • Kilauea Volcano, Hawaii · volcano, open event — touches this book
  • Sheveluch Volcano, Russia · volcano, open event
  • Suwanosejima Volcano, Japan · volcano, open event
  • Lewotobi Volcano, Indonesia · volcano, open event
  • Marapi Volcano, Indonesia · volcano, open event
  • Dukono Volcano, Indonesia · volcano, open event
  • Ibu Volcano, Indonesia · volcano, open event
  • Stromboli Volcano, Italy · volcano, open event
  • Aira Volcano, Japan · volcano, open event
  • Merapi Volcano, Indonesia · volcano, open event
  • Santa Maria Volcano, Guatemala · volcano, open event
  • Semeru Volcano, Indonesia · volcano, open event
  • Reventador Volcano, Ecuador · volcano, open event
  • M6.0 earthquake · 214 km NW of Port-Olry, Vanuatu — Oct 28, 02:21 UTC
  • M5.6 earthquake · 174 km NW of Port-Olry, Vanuatu — Oct 28, 02:59 UTC
  • M5.5 earthquake · 5 km ESE of Asahi, Japan — Oct 28, 15:07 UTC
  • M5.5 earthquake · Reykjanes Ridge — Oct 28, 11:37 UTC
  • M5.5 earthquake · 221 km NW of Port-Olry, Vanuatu — Oct 28, 01:39 UTC
  • M5.4 earthquake · 193 km WSW of Merizo Village, Guam — Oct 27, 23:14 UTC
  • M5.4 earthquake · 210 km ENE of Nagqu, China — touches this book — Oct 27, 18:57 UTC
  • M5.2 earthquake · 220 km NW of Port-Olry, Vanuatu — Oct 28, 02:17 UTC
  • M5.2 earthquake · 6 km NW of Balangonan, Philippines — Oct 28, 01:11 UTC
  • M5.1 earthquake · 7 km N of Balangonan, Philippines — Oct 27, 23:43 UTC
  • M5.0 earthquake · Reykjanes Ridge — Oct 28, 12:47 UTC
  • M5.0 earthquake · Reykjanes Ridge — Oct 28, 11:56 UTC
  • M5.0 earthquake · 205 km SSE of Vilyuchinsk, Russia — Oct 28, 06:12 UTC
  • M5.0 earthquake · 200 km W of Sola, Vanuatu — Oct 28, 04:31 UTC
  • M4.9 earthquake · 88 km SE of Hihifo, Tonga — Oct 28, 12:35 UTC
  • M4.9 earthquake · 1 km WSW of Butulan, Philippines — Oct 28, 12:14 UTC
  • M4.8 earthquake · 29 km SSW of Sarangani, Philippines — Oct 28, 09:32 UTC
  • M4.8 earthquake · 217 km NW of Port-Olry, Vanuatu — Oct 28, 02:13 UTC
  • M4.7 earthquake · 81 km ENE of Mazha, China — touches this book — Oct 28, 15:40 UTC
  • M4.7 earthquake · 39 km SSW of Serrezuela, Argentina — Oct 28, 07:02 UTC
  • M4.6 earthquake · south of the Fiji Islands — Oct 28, 12:58 UTC
  • M4.5 earthquake · 13 km WSW of Āwash, Ethiopia — Oct 28, 13:31 UTC
  • Satellite fire detections over 24 hours (NASA FIRMS): the 60 densest one-degree cells, 25,254 detections, the most in Botswana (767)
  • Illustrative traffic — sample positions: 30 ships on 7 shipping lanes and 11 flights between hub cities, not tracked
THE ENGINE

Markets don't
compile
.

Vibe coding works because code has a compiler — the machine tells you, instantly and without mercy, when you're wrong. Markets tell you slowly, noisily, and sometimes lie for years. Finquill is the missing toolchain for conviction: capture the thesis, arm its kill criteria, red-team it before you size it, monitor it while you live your life, and grade the record when reality reports back.

01CAPTURE

Say it in plain English. The engine structures it into falsifiable claims with deadlines.

02STRESS-TEST

A panel of adversarial agents tries to kill it. Four lenses; default verdict: refuted.

03MONITOR

Your kill criteria are checked against market data every week and around the company's earnings — or right away, on demand.

04RESOLVE

Claims come due. Reality grades them.

05RECORD

Brier-scored calibration plus process grades: was the decision right given what was knowable?

Example response
sample numbers

Thesis structured. 3 falsifiable claims: (1) top-3 hyperscaler capex guidance holds or rises through FY27 — resolve by each earnings; (2) Blackwell lead times stay >6mo through H1; (3) AI-accelerator mix >35% of DC revenue. Kill criteria armed — I'll flag guidance cuts >10% the morning they print.

Thesis captureFalsifiable claimsKill criteria
View ChartsFull ResearchGet Started
TRADING ON VIBES
A hunch from a headline
Conviction that hardens with the position
No record of why you entered
Hindsight rewrites the story
Zero learning per trade
REASONING WITH FINQUILL
A falsifiable thesis with a deadline
Red-teamed before it's sized
Kill criteria armed and watched
Graded on reasoning and outcome
Every trade teaches
Deep Read · real output

What a Deep Read finds.

Point the agent at one question and it reads the filings, writes a cited dossier, and stakes testable claims. These are real runs, published as written — each claim is graded against what actually happens.

LLYas of Oct 28, 2026

“$LLY — How durable is Lilly's lead in GLP-1 obesity drugs — supply, pricing pressure, the oral pipeline — and how much of it is already priced in?”

What consensus is missing
  • ▸Oral upside — Foundayo ~$98M in Q2 2026 vs injectable tens of billions; Novo oral already live.
  • ▸Net price path — Filings already show lower realized prices; Medicare discounts from Jul 1, 2026.
  • ▸Supply as moat — Shortages easing industry-wide; capacity is table stakes.
Claims staked · 4
  • At the Q3 2026 earnings release on 2026-10-29, Lilly will again characterize Mounjaro/Zepbound growth as driven primarily by increased volume and partially offset by lower realized prices, extending the FY2025 and H1 2026 pattern. open · by Oct 29, 2026
  • Lilly Q3 2026 total revenue will be at least $22.1 billion when reported on 2026-10-29, consistent with an intact volume-compounding tirzepatide franchise and the low end of the cited street range. confirmed
  • Foundayo (orforglipron) will not become a 2026–27 EPS savior at injectable scale: full-year 2027 Foundayo revenue will remain below $5 billion, in line with a slow 2027–30 oral build versus tirzepatide’s tens of billions. open · by Feb 28, 2028
  • Tirzepatide (Mounjaro + Zepbound) will still account for at least 50% of Lilly’s total revenue in FY2026, reflecting sustained—not diminishing—franchise concentration after 56% in FY2025 and 65% in H1 2026. open · by Feb 28, 2027
Read the full dossier
SPCXas of Oct 27, 2026

“$SPCX — How much of SpaceX's valuation rests on Starlink subscriber growth versus launch and AI, and what do its first public filings show?”

What consensus is missing
  • ▸Profit mix vs revenue mix: SI can be huge on revenue and small on profit ($431M vs $4.4B)—if true, market may overpay for SI revenue and under-appreciate…
  • ▸Launch/Starship real options not in the CFO soundbite—milestone re-rates can still surprise if cadence is buried.
  • ▸Tesla committee optionality is easy to overtrade; “no guarantee” means merge premium may be noise, not DCF.
Claims staked · 2
  • SPCX year-over-year revenue growth in the next earnings print will fall below the Q3 2026 rate of ~161%, as sustaining that print rate is over-appreciated and mean-reversion risk is material. open · by Feb 28, 2027
  • When the Q3 2026 10-Q segment disclosure is available, Super Intelligence will remain a minority share of consolidated profitability versus Connectivity/other, consistent with press figures of ~$431M SI profit vs ~$4.4B group NI. open
Read the full dossier
AAPLas of Aug 12, 2026

“AAPL — what is not being priced into the Services gross-margin story over the next two quarters? Read the latest 10-K/10-Q segment disclosures and recent filings, and say specifically what consensus appears to be missing.”

What consensus is missing
  • ▸Services GM already peaked sequentially and stopped expanding YoY.
  • ▸The highest-margin dollar in Services — search/default licensing — faces a live re-pricing regime under DOJ Google remedies
  • ▸Regulatory take-rate pressure is a slow grind already visible in growth mix (App Store model changes, gaming), not a binary future event
Claims staked · 4
  • Apple's Services gross margin will be below 75.5% in the next earnings report (FY2026 Q4). open · graded after the next report
  • Apple's Services gross margin will be below 75.0% in the earnings report after next (FY2027 Q1). open · graded after the next report
  • Apple's stock price will fall below $150 at any time before 2026-12-31. open · by Dec 31, 2026
  • Apple's stock price will decline by at least 10% over a rolling 30‑day window ending on or before 2026-12-31. open · by Dec 31, 2026
Read the full dossier

A Deep Read is a paid, minutes-long run. Composed research surfaces are the free path.

PLATFORM//SURFACES

Everything you need.
Nothing you don't.

Fifteen integrated surfaces on one reasoning engine. Built AI-native from the first commit.

Dashboard

What changed for what you hold and watch since you last looked — a thesis-delta digest, not a quote wall.

Thesis-delta view

Watchlist

Every name carries a living thesis — checked against market data weekly, rehearsed before its earnings print and autopsied after.

Weekly + earnings checks

Journal

A coach, not a diary — post-event interviews that separate your edge from your tilt.

Edge vs. tilt

Research

Cited theses with testable claims. Every number from a tool; every quote traceable to a source.

Citations everywhere

AIStudio

Eight specialists behind one conversation — an orchestrator plans, fans work out in parallel, and one voice synthesizes.

8 specialists

Trading

Preview → confirm on every order. Typed confirmation when it's live and the size is real.

Safety-gated

Monitor

The world on one lens, your exposure on the other — quakes, storms, chokepoints, sanctions, touching your book.

Dual-lens OSINT

Calendar

Earnings, dividends, and macro events — with your exposure and your theses attached.

Exposure-aware

Explore

Public dossiers, trending research, and the firm's own graded record.

Open door
15
Surfaces
10
AI Specialists
35
Tool Modules
PUBLIC
The record → /verify
AI//Investing in change

Notice the change. Then test whether the market has.

Great ideas often start as something you notice: a habit shifting, a product spreading, a company quietly buying differently. Finquill walks that observation through a disciplined loop — who it moves, how big it could get, what is already priced in — and holds you to a falsifiable thesis. The language model writes the sentence; the figures are meant to come from tools and cited sources.

01What's changing?
Step 1
Notice the shift
Behaviour, not headlines

Something you keep seeing: people or companies doing things differently. Say it in plain English, even half-formed.

02Who does it actually move?
Step 2
Trace it to the company
Materiality

A real shift only matters if it reaches a listed company's numbers. Finquill traces it and tells you how directly.

03How big could it get?
Step 3
Size it
Market, adoption pace, cost curve

Addressable market, adoption pace and cost curve, built from sourced figures with the sources shown. Where a figure can't be sourced, Finquill is built to say so rather than estimate.

04What's already priced in?
Step 4
Check expectations
What consensus expects

Compare what consensus already expects with your view, so you know whether the market has noticed yet.

05What would prove you wrong?
Step 5
Commit a falsifiable thesis
Claims, kill criteria, red-team

Your thesis becomes falsifiable claims with deadlines, then a red-team panel tries to break it before you size anything.

06Were you right, and how early?
Step 6
Grade your record
Calibration

Claims come due and reality grades them. Over time you see how well calibrated you are. Being early is the goal we're building toward.

Analytics, not advice — Finquill describes what the evidence shows and leaves the decision to you. Try the loop in AI Studio.

PROOF//OUR OWN BOOK

We run it on ourselves.
In public.

An autonomous research firm runs this same engine on its own coverage — publishing theses with kill criteria attached, changes of mind on the record, and a graded record anyone can check.

Scoreboard
0.18
Brier · 98 graded
Receipts
Every thesis update, call and grade is in a public dataset — misses included. Receipts are frozen at the moment they are minted.
Verify →Analytics, not advice.
RESEARCH ENGINE

Integrated research
environment.

Our intelligent RAG pipeline ingests SEC filings, earnings transcripts, news, and market data — then synthesizes it into actionable research you can query in natural language.

Multi-Source Ingestion

Continuously indexes SEC filings, earnings calls, analyst reports, and news into one queryable corpus.

Contextual Retrieval

Semantic search across your entire research corpus — ask a question and get answers grounded in real source documents.

AI Synthesis

Generates investment theses, competitive analyses, and risk assessments with full citation trails back to source material.

research / NVDA-thesis.mdCITED
Investment Thesis — structured for review

NVDA: data-center capex sustainable through FY27

Captured in plain English, then decomposed into claims that can actually be settled. Every figure is read out of a filing or transcript at generation time — a number the tools can't retrieve is refused, never estimated.

[1]
Top-3 hyperscaler capex guidance holds or rises
Resolves each earnings · 10-Q + call transcripts
[2]
AI-accelerator share of data-center revenue holds above its level at capture
Resolves quarterly · segment disclosures
[3]
Competing ASIC ramp does not compress realized ASPs
Resolves FY27 · your uploaded research
Kill criterion — armed
A combined guidance cut of more than 10% flags this thesis the morning it prints.
Every claim
Source-linked
AI//Prompt library

Start with what's changing. Then test it.

Finquill's own lenses spot a shift early, size the opportunity and check what the market already expects. The Market minds below are a separate set: classic frameworks to stress-test the same idea from another angle. Switch by tag, or stack several on one query.

Finquill lenses

Built for investing in change

Spot the Shift
Notice change early

Start from something you've seen change in how people or companies behave, then trace it to the listed companies it could actually move.

Example prompt
$I keep noticing more people paying for AI tools out of their own pocket. Which public companies would that actually move, and how directly?
Technology Adoption Sizing
Adoption, cost curve, market size

Size a new technology platform with sourced figures only: addressable market, adoption pace and cost curve. Figures carry their sources; where one can't be sourced, Finquill is built to say so rather than estimate.

Example prompt
$Size the market for humanoid robots over the next five years: addressable market, adoption pace and cost curve. Sourced figures only.
What's Priced In
Expectations vs. your view

Lay out what consensus already expects, then compare it with the thesis you're testing.

Example prompt
$What does consensus currently expect for NVDA, and where does my thesis differ from it?

Market minds

Classic frameworks distilled from famous investors' public writing — a second opinion on any idea

Buffett & Munger
Durable moats & quality

Screen the way Berkshire does — moat strength, owner earnings, returns on tangible capital.

Example prompt
$Score my watchlist on Buffett-Munger quality criteria; flag any name where the moat has narrowed in the last 4 quarters.
Howard Marks
Second-level thinking

Look past the consensus. What does everyone already know, and where is the non-obvious read?

Example prompt
$What's the consensus view on NVDA right now, and what's the second-level read a careful contrarian would build from the same facts?
Ray Dalio
All-weather & regime balance

Pressure-test portfolios against growth, inflation, and deflation regimes.

Example prompt
$Score my portfolio for factor balance across growth, inflation, and deflation regimes — flag any regime exposure above 40%.
Peter Lynch
Invest in what you know

Simple story, growing earnings, boring industry — the ten-bagger pattern, systematized.

Example prompt
$From the names on my watchlist, which match the Lynch 'simple story / growing EPS / under-covered' pattern? Rank by how closely they fit.
Ben Graham
Margin of safety

Defensive-investor criteria with today's data — quantitative, uncompromising.

Example prompt
$Run a Graham defensive-investor screen on my holdings. Flag any name outside his quantitative criteria with the specific line item that fails.

Build your own. Codify your investment philosophy as a reusable prompt once, then apply it to every new name you research.

Library grows every release
AI//Context × model fit

Your context, on every question.

Finquill enriches every query with your watchlist, economic calendar, and media feed. Choose Quick for speed or Smart for depth on each question, and pin the model you prefer for any agent in Settings.

Your watchlist + news feed
Model you can pin
Claude
Reasoning

Cross-portfolio implications of today's CPI print — reasoned through each of your names with their individual rate-sensitivity.

10-Ks + transcripts you've saved
Model you can pin
Gemini
Long context

Side-by-side comparison of the last four filings for three names you're researching — all in a single context window, nothing summarized away.

Quick or Smart per question; pin a model per agent in Settings.

AI//Model choice + frontier speed

You decide which model touches your work.

Open-source for full data sovereignty. Frontier closed-source for max capability. Diffusion models when you want ten draft theses to compare in seconds. Your call.

Your choice of model

Every major frontier model. Open and closed source.

Pin a specific model for a particular workflow in Settings. Default to the strongest available frontier for everything else, and choose Quick or Smart on each question.

OpenAI
Anthropic
Google
Meta · Llama
Mistral
xAI
Built for speed

Answers at the speed of thought.

We route to optimized inference compute (including Groq for ultra-low-latency models) so iterative research doesn't turn into a coffee break between every question. The faster the loop, the more questions you ask.

$ finquill --init

Stop trading on
vibes
.

Capture your first thesis in plain English. The engine makes it falsifiable, red-teams it, and starts your record. Capture is always free; red-team reports and deep research use credits, and you start with a free allowance. See pricing.

Free to start
Capture always free
Analytics, not advice
Early access — actively onboarding new users