P09 · Paper portfolio
Crypto 24/7 (EXPERIMENTAL)
by AI Portfolios
This is the simple view. Nothing has been removed — the decision records, the refused proposals, the AI's own words, the cost model, the published limits and the hash chain are all in Advanced, which is the switch in the header. How to verify any of it →
Showing every layer. Pick Simple or Advanced in the header to choose one — nothing is removed either way, and the full record is always in this page.
Crypto 24/7 (EXPERIMENTAL)
Nothing is due. Risk management — stops, forced exits and the NAV mark — runs on every pass regardless; only the research cycle waits for its published instant, which is 2026-09-26T17:00:00+00:00.
Test, in public and with realistic entry-tier retail costs, whether an LLM given delayed venue data can produce positive expectancy trading spot crypto on a 24/7 cadence, across weekends and through full crypto volatility regimes — over the whole venue-listed market above a published size floor, not over a hand-picked shortlist of majors (v2.0.0).
| Slot | Last due | Answered | Status |
|---|---|---|---|
d0100 0 1 * * * · America/New_York | 2026-09-26 05:00 | 2026-09-26 07:45 | ok due 2026-09-26T05:00:00+00:00, ran 2026-09-26T07:45:03.604253+00:00 |
d0700 0 7 * * * · America/New_York | 2026-09-26 11:00 | 2026-09-26 13:15 | ok due 2026-09-26T11:00:00+00:00, ran 2026-09-26T13:15:03.873406+00:00 |
d1300 0 13 * * * · America/New_York | 2026-09-25 17:00 | 2026-09-25 19:45 | ok due 2026-09-25T17:00:00+00:00, ran 2026-09-25T19:45:03.841867+00:00 |
d1900 0 19 * * * · America/New_York | 2026-09-25 23:00 | 2026-09-26 01:45 | ok due 2026-09-25T23:00:00+00:00, ran 2026-09-26T01:45:03.619109+00:00 |
Catch-up policy. A decision instant that passes unanswered is run on the next pass, once, and the record stamps both instants — when it was due and when it actually ran. Past the catch-up window the gap stays open and is published as missed rather than backfilled.
Track record
Fewer than 60 daily observations. Annualised return, volatility, Sharpe, Sortino and Calmar are withheld: annualising a short sample produces numbers that look precise and are not.
Our own marks from our own fills (nothing here is vendor price data). Deliberately monochrome: colour that codes "good" and "bad" is an editorial thumb on the scale.
no benchmark series for BTC_SPOT over this window; the portfolio line is shown alone rather than against a different period
The line is rebased to 100 at the first mark in the record. No benchmark is drawn, so nothing here is a comparison.
Index values are derived percentage returns between our own recorded observations; no raw vendor price is rendered.
| symbol | since entry | weight | opened |
|---|---|---|---|
| LINK | +54.3% | +15.8% | 2026-08-14 |
| MORPHO | +34.5% | +9.3% | 2026-08-15 |
| CRV | +23.9% | +7.9% | 2026-08-14 |
| ETH | +42.7% | +7.5% | 2026-08-16 |
| TAO | +67.6% | +6.6% | 2026-08-17 |
| SOL | +56.1% | +4.1% | 2026-08-18 |
| BTC | +29.8% | +1.6% | 2026-08-17 |
| ETHFI | +45.1% | +1.6% | 2026-08-14 |
| ZEC | +210.1% | +1.3% | 2026-08-15 |
Return since the AI entered, measured from this book's own mark against what it actually paid — our own numbers over our own record, not vendor price data.
Weights are our own work product, computed at the newest audit-chain NAV mark — never re-derived later, never estimated from cost basis. Logos identify the holdings and imply no endorsement (operator decision, session 11); sector tints are structure, not judgement.
| Symbol | Quantity | Avg cost | Stop | Opened |
|---|---|---|---|---|
 BTC1.6% | 0.0250 | 64731.74 | 61600.50 | 2026-08-17 |
| CRV7.9% | 29399.8979 | 0.28 | 0.28 | 2026-08-14 |
 ETH7.5% | 3.5751 | 1884.29 | 1770.61 | 2026-08-16 |
| ETHFI1.6% | 2875.5073 | 0.50 | 0.44 | 2026-08-14 committed 2026-08-15 |
 LINK15.8% | 1438.7233 | 9.21 | 8.64 | 2026-08-14 |
| MORPHO9.3% | 4321.0911 | 2.05 | 1.99 | 2026-08-15 |
 SOL4.1% | 44.0934 | 77.56 | 72.55 | 2026-08-18 |
| TAO6.6% | 25.7842 | 197.68 | 176.76 | 2026-08-17 |
| ZEC1.3% | 1.1122 | 497.36 | 434.62 | 2026-08-15 |
Why two dates? "Opened" is the market session the opening fill was priced on. The licensed data feed's T+1 rule means a cycle prices on the newest session already licensed for use, so the run that produced the fill can be committed to the audit chain on a later date — that commitment date is shown beneath. The commitment time is what makes history tamper-evident; the pricing session is what returns are computed from. Both are published on every fill.
Recent decisions
Every run, including the ones that traded nothing: the committed fact-pack hash (written to the audit chain before anything was proposed), the AI's published view, each proposal with the rule engine's verdict, and the fills that resulted. Rejected proposals stay up with their reason codes — a reader who can only see the trades that happened cannot judge the discipline that refused the ones that didn't.
Every card carries the instant it was decided at, because every card is a photograph. The facts in it were frozen when the run was committed and are never revised — a view that said "the platform has closed zero trades" was true at its timestamp and reads as a claim about today unless the card says otherwise. Consecutive runs that halted on identical codes are compressed into one card with a visible count and every run id listed; nothing is dropped.
What the AI is allowed to do
This portfolio's AI ranks candidates and explains its reasoning. It proposes direction and conviction only. Every price, quantity, weight and risk limit is computed by deterministic code the model cannot see or influence.
BUY · ADD · TRIM · SELL · HOLD · TIGHTEN_STOP
At most 3 proposals per run, each carrying a direction, a conviction from 1 to 5, a rationale, and references to the facts it used.
- set a price
- set a quantity or share count
- set a weight or position size
- set or move a risk limit
- execute anything
Model tier fast · prompt version P09-v1. Any change to the decider or the prompt is recorded as a MODEL_CHANGE event in the audit chain — without it the record would be a chimera of two systems presented as one.
Enforced limits
Every limit below names the code path that enforces it. This is not decoration: a limit that is published but never fires is worse than no limit, because a reader has no way to tell the difference. A test walks every mandate, drives the engine into the state each declared control claims to protect against, and fails the build if nothing stops it.
| Limit | Value | Enforced by |
|---|---|---|
| Maximum gross exposure | 95.0% | gate 2 — size_order |
| Maximum single sector | 100.0% | gate 2 — size_order |
| Limit | Value | Enforced by |
|---|---|---|
| Shorting | prohibited | gate 1 — proposal validation |
| Derivatives | prohibited | gate 1 — proposal validation |
| Margin | prohibited | gate 1 — proposal validation |
| Leverage | prohibited | gate 1 — proposal validation |
| Limit | Value | Enforced by |
|---|---|---|
| Maximum positions | 100 | gate 3 — BREACH_MAX_POSITIONS (portfolio limit check) |
| Minimum new position | 0.5% | gate 2 — size_order |
| Maximum order as % of ADV | 0.5% | gate 2 — size_order |
| Limit | Value | Enforced by |
|---|---|---|
| Daily loss limit | 4.0% | gate 3 — HALT_DAILY_LOSS_LIMIT (halts and flattens) |
| Weekly loss limit | 10.0% | gate 3 — HALT_WEEKLY_LOSS_LIMIT |
| Monthly loss limit | 20.0% | gate 3 — HALT_MONTHLY_LOSS_LIMIT |
| Limit | Value | Enforced by |
|---|---|---|
| Stale-data halt | 2 hours | gate 3 — HALT_STALE_DATA |
| Limit | Value | Enforced by |
|---|---|---|
| At −20.0% drawdown | alert, publish_commentary | gate 3 — drawdown ladder |
| At −35.0% drawdown | alert, scale_risk_budget:0.5, pause_new_positions_days:5 | gate 3 — drawdown ladder |
| At −50.0% drawdown | suspend_portfolio, publish_post_mortem | gate 3 — drawdown ladder |
| At −70.0% drawdown | graveyard_review | gate 3 — drawdown ladder |
| Limit | Value | Enforced by |
|---|---|---|
| Max round trips per week | 15 | entry block — PAUSE_TURNOVER_CAP_REACHED (never blocks an exit) |
| Max trades per month | 40 | entry block — PAUSE_TURNOVER_CAP_REACHED (never blocks an exit) |
| Limit | Value | Enforced by |
|---|---|---|
| Halt trading if price gap unexplained pct | 40 | gate 3 — circuit breaker |
| Halt trading if venue price divergence bps above | 300 | gate 3 — circuit breaker |
| Halt trading if factpack hash mismatch | true | gate 3 — circuit breaker |
Execution
Fills are struck on a VWAP window that begins strictly after the decision timestamp plus a latency delay. No fill can use information from the bar in which the decision was made, so look-ahead is arithmetically impossible rather than merely avoided.
Philosophy
Crypto never closes, so neither does the risk. This portfolio holds positions overnight and through weekends — unlike the Intraday AI Lab it is a swing book on a continuous market, not a flat-by-close experiment. Since v2.0.0 it is also the platform's widest book by name count: the operator's judgement is that ten hand-picked coins is not a crypto portfolio, so the universe is every asset the venue lists above a size floor and the selection is the experiment. What remains constrained is loss per period and what may be OWNED — deterministic per-asset caps, mandatory stops sized off realised range, and a volatility regime the AI cannot override. Spot only. No leverage, no staking, no yield, no wrapped or synthetic assets, no stablecoins, and no memecoin without two years of history.
Review schedule
| Job | Cron | Timezone | Makes a decision |
|---|---|---|---|
| d0100 | 0 1 * * * | America/New_York | yes |
| d0700 | 0 7 * * * | America/New_York | yes |
| d1300 | 0 13 * * * | America/New_York | yes |
| d1900 | 0 19 * * * | America/New_York | yes |
Published on a fixed, regular schedule. Never event-triggered — no "market is crashing, sell now" alerts, deliberately. This table is the promise; the cadence panel at the top of the page is whether it was kept, instant by instant.
Machine-readable mandate: config/portfolios/P09_crypto_24_7.yaml · human rulebook: docs/portfolios/P09_crypto_24_7.md · effective from 2026-08-10 (wall clock) · record starts 2026-08-14 (pricing session) · payload 540d30e63326…