War desk — LTC joint paper trial — readout 2026-08-08 · paper only · honest updates

Get the weekly desk note

Governed paper desk · multi-engine

We don't sell trades.
We sell a desk honest enough to know when not to.

Ops software for paper trading — not a signal service, not guaranteed returns, not another “AI tips” bot. Built to squash scam trading sites with process you can audit.

Learning Core closed NO_EDGE on SOL — Active is the LTC war desk now.

Skip taxonomy, fleet truth, and an audit ledger for serious paper traders. No black-box AI entries. No guaranteed returns.

PAPER DEFAULT · OWNER AUTHORITY · NET AFTER FEES

Rook Hybrid Core — human judgment fused with machine honesty
Hybrid Core™ — human authority · machine honesty

What we aim for — proven in paper before any live capital conversation. Not a get-rich pitch.

Sustainable net profit after fees

Across a portfolio of independent trading engines — crypto SPOT paper first, then additional asset classes as separate engines, not one monolithic bot.

Multi-engine profit after fees — owner-first, paper-proven, asset-class expandable.

Advance and learn is the new growth.

Rook runs independent paper engines (crypto now; equities and others later), proves net after fees before any live capital conversation, and treats correct silence as success — not a bug to market away.

Viability ladder GATHERING

Live rollup from paper desk — not a marketing claim. Profit evidence gates every tier.

  1. LOCKED Machinery or calibration not ready
  2. GATHERING Honest machinery — gathering fee-aware paper evidence
  3. CRYPTO_PAPER_VIABLE Primary engine 21d net + calibration complete
  4. MULTI_ENGINE_PAPER_VIABLE Two+ engines 21d net — portfolio lane
  5. LIVE_POLICY_CANDIDATE Owner may discuss live policy in writing — never automatic

Operating principles

  • Paper before capital — proof in paper until you change policy in writing.
  • You are final authority — automation may tighten or explain, never silently loosen.
  • Edge over activity — net after fees beats fill count; skips are often correct.
  • One focus at a time — one Active hub task (LTC war desk Long + refuse now).
  • Legible outcomes — what happened, why, how much, what next — without reading code.
  • Multi-engine profit — portfolio of engines, not one hero bot forever.

Why the focus changed

The turnaround

The Great Review

Hard decision. Right decision. We paused the 21-day Day-N ceremony on purpose — not because the method failed, but because process alone is not enough. Learning Core on SOL returned NO_EDGE — kill accepted. Public story tracks the LTC war desk next.

No coin-hop. No gate loosen to force a story. Paper engines stay running as the label source.

Past

We productised a 21-day calibration cycle on LTC — truth week, one knob, Day 18 keep/revert, sign-off. That method shipped. LINK ceremony is frozen archive.

Pivot

The Great Review — pause the Day-N clock to build a Learning Core (feature export, walk-forward harness, kill criteria) that must show EDGE or accept an honest negative.

Now

SOL Learning Core gather closed with honest NO_EDGE (vaulted). Active clock is LTC Long + refuse war desk — readout 2026-08-08. One Active hub task. Observe peers; don't chase fills.

Horizon (12–24 mo): Honest harness EDGE (or stop) → multi-engine 21d net → TradFi as separate engines → SaaS / SILENCESTORM only after that honesty gate. Live capital is a written owner decision, not a launch milestone.

P2

Calibration method (shipped)

LTC cycle complete; LINK ceremony paused as archive. Method stays — live clock does not.

P3

War desk (active)

LTC Long + refuse through 2026-08-08. SOL LC = NO_EDGE closed. Trail peers observe-only. Paper only.

P2b

Multi-engine paper profit

Two+ crypto engines with honest 21d net after fees; parity and metrics snapshots PASS.

P4

Scale + new asset classes

Per-engine independence, TradFi charter, portfolio without silent concentration.

P5

Production hardening

Job ledger, alerts, Safeguard apply — still owner-first, still paper default.

Explicitly not on the roadmap

  • Silent gate loosening to force fills
  • Live capital without written owner policy
  • Black-box models without walk-forward PASS
  • Signal feeds and copy-trading malls as primary edge
  • Guaranteed-return marketing or AI entry picks

What we're proving, in order

Every goal has a PASS/FAIL artifact — program board, viability rollup, case study, or peer script. We do not skip ahead because the calendar says so.

Near term Now → war desk readout

  • LTC war desk Long + refuseScore through 2026-08-08 · no TQ loosen
  • Trail-primary peers BTC/LINK/SOLObserve only · early 08-08 · full 08-17
  • Keep paper desk machinery greenFleet RUNNING · Morning/EOD · no marketing env edits
  • SOL LC = NO_EDGE closedVaulted · not Active · no coin-hop

Medium term After honesty

  • Case study + Truth Certificate™Artifact coherence — not PnL hype
  • CRYPTO_PAPER_VIABLE held honestly21d net + parity + Learning Core honesty
  • Operator tier betaPaper SaaS · waitlist → Stripe when honest

Long term 12–24 mo

  • MULTI_ENGINE_PAPER_VIABLETwo+ engines · portfolio north star
  • TradFi paper laneSeparate engine pattern · shared ops
  • SILENCESTORM™ + peer GTMAfter honest EDGE — advisory-only

Built for operators burned by green lies

Primary — semi-pro paper traders

Run multiple bots or strategies seriously. Hate contradictory dashboards, silent gate drift, and platforms that celebrate activity over net after fees. Will read skip logs on a Saturday morning.

Secondary — Freqtrade / Jesse builders

Keep your strategy stack. Rook supplies Morning/EOD spine, fleet truth, calibration product, and owner ledger — the ops shell you would otherwise duct-tape together.

Not for

Get-rich-quick retail, copy-traders wanting Telegram signals, or anyone who needs guaranteed returns on a landing page. We will send you away politely.

What scam platforms optimise — and what we refuse

Mainstream bot platforms sell activity and excitement. Rook optimises trust, patience, and net after fees across independent engines. The hero metric is often correct silence.

We celebrate skips

Every blocked trade is explained — reason code, fee multiple, quality gate. Not "missed opportunity."

One knob, one backup

Calibration cycles lock changes. Apply verified on disk. Revert from .bak — logged in the owner ledger.

Same number everywhere

Metrics snapshot + parity checks. If Coin and Elite disagree, the desk FAILs loud — not quietly.

We never ship

AI entry picks · silent gate loosening · guaranteed ROI marketing · copy-trading signal malls.

Not a better bot. A better desk.

Freqtrade and Jesse are excellent for strategy dev. 3Commas and Cryptohopper sell activity. Rook owns the gap: operations, truth, and calibration.

Capability Typical bot platform Rook
Why no trade? Dig per-bot event logs Skip taxonomy + Doctor + plain English
Multi-bot fleet Many bots, no governance Fleet spine + entanglement alerts
Calibration Ad hoc tuning 21-day productised cycle
Proof artifact PnL screenshot Truth Certificate™ + case study
AI role Pick entries / predict Explain skips · audit · never auto-loosen
Scheduled ops spine Cron + DIY health checks (Schduler-class) Morning Ops + EOD sync + Patience Score™ — owner truth, not just uptime
Verify / FAIL honesty PnL screenshots · deleted losing weeks Truth Certificate™ verify URL + Death Certificate™ export
Hype / trending tape Chase narrative · rotate to hot coins SILENCESTORM™ Refusal Intelligence — PAUSE/RUN/FOCUS (roadmap)
Authority Platform vendor Owner — paper default, explicit live policy

The Correct Silence Index™ is the product

Fake algos celebrate fills. Rook explains skips with reason codes, Doctor, and fleet entanglement — silence after a verified knob tighten is often success.

Correct Silence Index™: 100% skips · quality / fee gates blocked — gates doing their job.

Patience Score™: machinery trust — not PnL rank.

Parity badge: Coin and Elite must agree — or the desk FAILs loud.

ROADMAP · SS-01 after honest EDGE

SILENCESTORM™

When hype storms, your desk refuses with proof.

SILENCESTORM™ scores Narrative–Machinery Divergence™ (NMD) when trending tape diverges from your skip taxonomy — and outputs PAUSE · RUN · FOCUS for your attention, never for capital. Advisory-only; not signals; not live until after honest Learning Core EDGE (ceremony paused — calendar Day 21 alone is not the gate).

Correct Silence Index™ (CSI)

Is silence correct for this tape — or a machinery fault? Regime + skips + fleet correlation fused.

Attention Receipt™

Weekly export: coin trended · desk refused · owner chose PAUSE — extends Truth Certificate™ and Death Certificate™ trilogy.

Epoch lock

Advice reads your Active focus and calibration contract — no FOCUS that implies loosen mid-gather.

Never

Auto entries · gate loosen · copy-trade · “put money in X” · performance hero marketing.

Batteries-included ops for paper traders

Morning Ops, EOD sync, program board, and owner gates — the machinery solo traders duct-tape together on Freqtrade. Included from day one.

Elite desk truth

One headline verdict: machinery, economics, Learning Core focus. Back-seat in ≤60 seconds on observe days.

Coin + Fleet

Per-engine live state, why-no-trade panel, process controls. Fleet health without tuning other coins mid-cycle.

21-day calibration

Shipped method: hypothesis → backup → apply → monitor → Day 18 keep/revert. Honest narrative — not today's live clock.

Learning Core

Feature export, walk-forward harness, kill date. Prove EDGE or stop — no learned gates until honesty.

Truth Certificate™

Exportable attestation: paper-only, parity-checked numbers, public verify page. Share proof, not hype.

Owner decision ledger

Append-only audit: apply, Day 18, sign-off, gap events. Your desk has memory — scammers prefer amnesia.

Adversarial AI

Portfolio narrator and ops copilot grounded in artifacts. Trade-leak CI. AI that audits, not trades.

21-day cycle = the method · war desk = today's clock

We productised a 21-day calibration cycle on LTC — truth week, one knob, Day 18 keep/revert, Day 21 sign-off. That method stays. The live Day-N ceremony is paused. SOL Learning Core closed with honest NO_EDGE. Active public clock: LTC war desk Long + refuse through readout 2026-08-08 (peers trail-primary observe-only).

Now

War desk · LTC Long + refuse

Score fee-honest KEEP/REVERT vs baselines through 2026-08-08. No TQ/fee/size loosen mid-trial.

08-08

Joint readout + peer early

LTC joint scoreboard · trail peers BTC/LINK/SOL early check (observe only — do not merge).

08-17

Peer full readout

Trail-primary full window. Still observe-only. Postures/sizing review only after owner gate.

Archive — how a formal calibration cycle looks when we run one again:

D1–7

Truth week

No instance.env edits. Fix stale machinery. Classify skip taxonomy. Baseline locked.

D8–10

Hypothesis

Peer compare (observe-only). Pick one allowlisted knob. Document before touch.

D11–14

Apply + monitor

Backup · apply · verify on disk. No second knob. Post-apply silence expected after tighten.

D18–21

Decision + sign-off

Keep / revert / observe vs frozen baseline. Case study rebuild. Then — and only then — peer GTM.

Process visible · knobs private

Sanitised from the live production hub at EOD — phases and titles only. No gate values, no instance.env, no raw fills.

Now

  • War desk · LTC Long + refuse
  • Score Long+refuse through readout 2026-08-08 — KEEP/REVERT · no TQ loosen
  • Trail-primary peers BTC/LINK/SOL — observe only (early 08-08 · full 08-17)
  • SOL Learning Core = NO_EDGE closed (vault) — not Active
  • Paper only · owner authority · net after fees

Next

  • War desk readout — KEEP / REVERT / observe2026-08-08 · fee-honest baselines
  • Trail-primary peers early checkEarly 2026-08-08 · observe only
  • Trail-primary peers full readoutFull 2026-08-17

Later

  • SILENCESTORM™ Refusal IntelligenceSS-01 after honest EDGE · advisory-only
  • Operator tier beta — governed paper deskAfter peer validation
  • Multi-tenant SaaS shellPost MULTI_ENGINE paper viable

Not on this site (by design)

    Connected to the desk you're running

    Hero metrics, proof block, and this list refresh from live paper ops — same artifacts as Elite and EOD. Static marketing copy stays; numbers and ship log track development.

    War desk · LTC Long + refuse

    Readout 2026-08-08 · SOL LC NO_EDGE closed · paper only · no TQ loosen

    Recently shipped

      Discoveries & breakthroughs

        Truth Certificate™ — sample (paper desk attestation)

        Scam platforms show PnL screenshots. Rook exports a signed desk attestation: machinery, metrics, parity — not a promise of profit. Sample from live paper ops (war desk · LTC · viability from desk).

        PASS rook-paper-…

        Desk attestation PASS — artifacts coherent. Not investment advice.

        Desk truthOK
        Metrics parityOK
        Viability rollup
        FocusWar desk · readout 2026-08-08
        Knob / ceremonyCeremony paused
        Top skip
        War desk: score LTC Long + refuse through 2026-08-08. SOL Learning Core = NO_EDGE closed (vaulted). Calibration Day-N is frozen archive, not today's clock.

        Verify online: verify/index.html · enter certificate ID from export. Full HTML: exports/truth_certificate_20260621.html in repo. Print → PDF for peer review (G-052).

        Subscribe to a governed desk — not signals

        Tier shapes below are the product direction. Prices are TBC until paper proof (viability + case study + peer validation). All tiers paper-only by default. Join the weekly desk note — we will not take payment before that gate.

        Trial

        TBC / 14 days

        • 1 paper engine
        • Elite + Coin lite
        • Ops runbook
        Join waitlist

        Desk

        TBC / mo

        • 10 engines · Fleet ops
        • Truth Certificate™ exports
        • Calibration program
        • Priority support
        Join waitlist

        Price set after honest paper proof — not before. Live execution: opt-in only, BYO keys, separate agreement. Rook never holds your capital.

        Hard questions — honest answers

        Is this financial advice?
        No. Rook is paper-trading operations software. We explain machinery and gates — not what to buy or sell.
        Do you guarantee returns?
        Never. We show net after fees, skips, and failure modes. Our Death Certificate™ export documents honest FAILs — see Elite / exports.
        How is AI used?
        Adversarial ops copilot and narrators read artifacts you already have. No auto-loosening. No entry generation. CI blocks trade-leak prompts.
        Why paper first?
        Because scam platforms blur paper and live to sell hope. We prove desk honesty in paper epochs before asking for money or live keys.
        Who is this for?
        Serious hobby / semi-pro quant traders who hate contradictory dashboards and will read skip logs. Not get-rich-quick retail or copy-traders.
        What is the north star?
        Sustainable net profit after fees across a portfolio of independent engines — crypto paper first, other asset classes as separate engines later. One coin calibration is a lane, not the finish line.
        When do you charge money or go live?
        Paid Operator tier only after paper proof artifacts (viability rollup, case study, peer validation). Live execution is opt-in, BYO keys, separate agreement — never default onboarding.
        Are you raising investment?
        No. TTC Holdings Group Ltd is owner-operated. Waitlist is for operator early access — not investor applications. Bootstrap via paper SaaS when proof exists.
        How does Kraken fit in?
        BYO paper API keys on your runner — we do not hold funds or trade custodially. Rook is the ops layer (skips, fleet, calibration), not a signal mall. Full FAQ →
        Do I need my laptop on 24/7?
        Engines need a host (your PC or VPS). The public site is moving to always-on hosting so visitors never depend on your laptop. Architecture →

        How Rook + Kraken works — full FAQ

        Get the weekly desk note.

        One email list: a short Friday process note (war desk honesty, skips, ships) — plus Operator early access when the paper proof is real. Not a signal service.

        First desk note targeted 31 Jul 2026 (owner-reviewed draft — not auto-sent).