Run frozen continuous capital studies and immutable revision pairs

.github/workflows/ci.yml

@@ -69,3 +69,6 @@ jobs:
         run: |
           python scripts/run_synthetic_demo.py --out "$RUNNER_TEMP/synthetic-demo"
           grep -q "SYNTHETIC DEMONSTRATION" "$RUNNER_TEMP/synthetic-demo/metrics.json"
+
+      - name: Continuous capital and immutable revision CLI acceptance
+        run: python scripts/run_continuous_demo.py --out "$RUNNER_TEMP/continuous-demo"

README.md

@@ -1,5 +1,8 @@
 # Quant Research Portfolio
 
+[Continuous shared capital and paired revisions](docs/CONTINUOUS_CAPITAL.md) add
+a runnable frozen-policy study on QExec's shared account and QDK's immutable lineage.
+
 A small, reviewable portfolio of quantitative-research engineering patterns:
 
 - deterministic synthetic data generation;

docs/CONTINUOUS_CAPITAL.md

@@ -0,0 +1,72 @@
+# Continuous shared capital and paired revisions
+
+Install this repository with its pinned QDK/QExec dependencies. The new commands
+are exposed by the existing `quant-portfolio` CLI. The deterministic example is
+entirely synthetic; it establishes accounting and revision plumbing only.
+
+```bash
+python scripts/run_continuous_demo.py --out /new/empty/path/capital-demo
+```
+
+This executes the real freeze, continuous-study and revision-study CLIs, pins both
+market versions and result derivations in QDK's `ArtifactStore`, and writes
+`acceptance.json`, `README.md`, both account reports, and a revision report. The
+independent scalar expectation is +34 USD old PnL, +26 new PnL, and -8 difference.
+No data, accounts, models, credentials or real orders are acquired or modified.
+
+Run individual steps with your own explicitly declared research inputs:
+
+```bash
+quant-portfolio freeze-capital-plan --plan examples/continuous/plan.json --out plan-frozen.json
+quant-portfolio continuous-study --plan plan-frozen.json --market examples/continuous/market-old.json --out old.json
+quant-portfolio revision-study --request request.json --out revision.json
+```
+
+The draft `quant.capital-plan/v1` carries full QDK `InstrumentSpec` payloads,
+instrument-to-sleeve ownership, opening currency allocations and FX, a fee-per-unit
+cost policy, valuation policy, decisions, and optional explicit cash events.
+Each decision fixes a notional target, signal observation, later execution
+observation and settlement instant. It rounds the prior-observation notional to
+quantity steps, then submits the target delta through QExec at the explicit later
+price. It does not adapt or optimize the policy to evaluation outcomes. Missing
+prices, unavailable observations, ambiguous order, insufficient funds, unsupported
+assets or silently changed controls fail the run. Executions are declared research
+fills, with explicit fees; this adapter does not claim order-book matching.
+
+Market input `quant.capital-market/v1` has observations with unique IDs, UTC `at`,
+`available_at <= at`, price mapping and FX mapping. Cash-event sequences occupy
+0–99, observation marks 100–999, decisions 1000 upward. The QExec account retains
+all ordered events, settlements, sleeves, journal postings and cash/margin state
+continuously. See QExec `docs/SHARED_CAPITAL.md` for the exact event contract and
+limitations. B's forecast replay may call that additive API with its own fills.
+
+Freezing copies the controls and binds actual installed package Python source
+hashes and runtime versions. Editing the plan, source or runtime requires a new
+explicit freeze and cannot be reported as a data-only comparison. Freezing is
+performed now: historical fixture dates do not confer historical/OOS provenance.
+The adapter supports `synthetic/deterministic_fixture` or explicitly declared
+retrospective studies, and exposes features/forecasts as unavailable.
+
+For revision requests, publish old/new market JSON using `ArtifactStore.put_bytes`
+and put their complete `ArtifactRef.to_dict()` objects in `old_input` and
+`new_input`. `request.json` also contains absolute `store`, frozen `plan` path and
+durable `owner`. `paired_study` checks the same frozen executable and controls,
+recalculates both outputs, and registers exact dependencies. The report includes
+transitive affected objects, signals/orders/history/NAV/return changes, unchanged
+stage results, original references and explicit `intervention=data_revision`.
+It never replaces earlier decisions. Dependency reachability is not causal proof.
+
+Reports emit the common `quant.measurement/v1`. Return observations group all
+same-instant events after respecting their explicit sequence; their return is the
+product of each flow-adjusted event return. The timestamp grid includes opening
+and final boundaries, is UTC and strictly increasing. Event returns are never
+annualized. Cash is in base-currency major units, returns are decimal fractions,
+and exact source, market, instrument, FX and economic-policy hashes are retained.
+Old/new revisions intentionally have different input identities and therefore
+must fail ordinary same-input economic comparison. The revision report is the
+explicit intervention analysis.
+
+Every CLI output uses exclusive creation. Keep frozen plans, raw references and
+artifact-store pins with the reports. Re-running into a different new directory
+is safe; overwriting past research is rejected. Roll back through Git revert while
+preserving output history. No deployment or default-branch merge is performed.

examples/continuous/market-new.json

@@ -0,0 +1,35 @@
+{
+  "schema": "quant.capital-market/v1",
+  "observations": [
+    {
+      "observation_id": "o1",
+      "at": "2026-01-02T00:00:00Z",
+      "available_at": "2026-01-02T00:00:00Z",
+      "prices": {
+        "A": "125",
+        "B": "50"
+      },
+      "fx": {}
+    },
+    {
+      "observation_id": "o2",
+      "at": "2026-01-03T00:00:00Z",
+      "available_at": "2026-01-03T00:00:00Z",
+      "prices": {
+        "A": "101",
+        "B": "50"
+      },
+      "fx": {}
+    },
+    {
+      "observation_id": "o3",
+      "at": "2026-01-04T00:00:00Z",
+      "available_at": "2026-01-04T00:00:00Z",
+      "prices": {
+        "A": "110",
+        "B": "50"
+      },
+      "fx": {}
+    }
+  ]
+}

examples/continuous/market-old.json

@@ -0,0 +1,35 @@
+{
+  "schema": "quant.capital-market/v1",
+  "observations": [
+    {
+      "observation_id": "o1",
+      "at": "2026-01-02T00:00:00Z",
+      "available_at": "2026-01-02T00:00:00Z",
+      "prices": {
+        "A": "100",
+        "B": "50"
+      },
+      "fx": {}
+    },
+    {
+      "observation_id": "o2",
+      "at": "2026-01-03T00:00:00Z",
+      "available_at": "2026-01-03T00:00:00Z",
+      "prices": {
+        "A": "101",
+        "B": "50"
+      },
+      "fx": {}
+    },
+    {
+      "observation_id": "o3",
+      "at": "2026-01-04T00:00:00Z",
+      "available_at": "2026-01-04T00:00:00Z",
+      "prices": {
+        "A": "110",
+        "B": "50"
+      },
+      "fx": {}
+    }
+  ]
+}

examples/continuous/plan.json

@@ -0,0 +1,113 @@
+{
+  "schema": "quant.capital-plan/v1",
+  "evidence_kind": "synthetic",
+  "sample_kind": "deterministic_fixture",
+  "instruments": [
+    {
+      "instrument_id": "A",
+      "asset_class": "etf",
+      "product_type": "cash_etf",
+      "venue": "SYNTHETIC",
+      "native_symbol": "A",
+      "base_currency": null,
+      "quote_currency": null,
+      "settlement_currency": "USD",
+      "price_tick": {
+        "units": 1,
+        "scale": 2
+      },
+      "quantity_step": {
+        "units": 1,
+        "scale": 0
+      },
+      "contract_multiplier": {
+        "units": 1,
+        "scale": 0
+      },
+      "calendar_id": "explicit_fixture",
+      "margin_mode": "none",
+      "inverse": false,
+      "effective_from": "2026-01-01T00:00:00Z",
+      "effective_to": null,
+      "available_at": "2026-01-01T00:00:00Z",
+      "superseded_at": null,
+      "underlying_id": null,
+      "expiry_date": null,
+      "metadata": {}
+    },
+    {
+      "instrument_id": "B",
+      "asset_class": "etf",
+      "product_type": "cash_etf",
+      "venue": "SYNTHETIC",
+      "native_symbol": "B",
+      "base_currency": null,
+      "quote_currency": null,
+      "settlement_currency": "USD",
+      "price_tick": {
+        "units": 1,
+        "scale": 2
+      },
+      "quantity_step": {
+        "units": 1,
+        "scale": 0
+      },
+      "contract_multiplier": {
+        "units": 1,
+        "scale": 0
+      },
+      "calendar_id": "explicit_fixture",
+      "margin_mode": "none",
+      "inverse": false,
+      "effective_from": "2026-01-01T00:00:00Z",
+      "effective_to": null,
+      "available_at": "2026-01-01T00:00:00Z",
+      "superseded_at": null,
+      "underlying_id": null,
+      "expiry_date": null,
+      "metadata": {}
+    }
+  ],
+  "owners": {
+    "A": "alpha",
+    "B": "beta"
+  },
+  "allocations": {
+    "alpha": {
+      "USD": "600"
+    },
+    "beta": {
+      "USD": "400"
+    }
+  },
+  "base_currency": "USD",
+  "opened_at": "2026-01-01T00:00:00Z",
+  "opening_fx": {},
+  "cost_policy": {
+    "model": "explicit_fee_per_unit",
+    "fee_per_unit": "1"
+  },
+  "valuation_policy": {
+    "model": "last_available_mark_and_fx",
+    "positive_futures_pnl_spendable": false
+  },
+  "decisions": [
+    {
+      "decision_id": "buy-a",
+      "signal_observation_id": "o1",
+      "execution_observation_id": "o2",
+      "instrument_id": "A",
+      "target_notional": "500",
+      "settles_at": "2026-01-03T00:00:00Z"
+    },
+    {
+      "decision_id": "buy-b",
+      "signal_observation_id": "o1",
+      "execution_observation_id": "o2",
+      "instrument_id": "B",
+      "target_notional": "300",
+      "settles_at": "2026-01-03T00:00:00Z"
+    }
+  ],
+  "cash_events": []
+}

pyproject.toml

@@ -10,8 +10,8 @@ readme = "README.md"
 license = { text = "MIT" }
 requires-python = ">=3.10"
 dependencies = [
-    "quant-data-kit @ git+https://github.com/PureSaber/quant-data-kit.git@ba136c2fa2eea121bfb2ad7887b536c3952586f7",
-    "quant-execution @ git+https://github.com/PureSaber/quant-execution.git@21aace45d2dde6458db8fcda4829f252f8c640f1",
+    "quant-data-kit @ git+https://github.com/PureSaber/quant-data-kit.git@f85128307dd0e3407d24350282fe7b679b591309",
+    "quant-execution @ git+https://github.com/PureSaber/quant-execution.git@49ef53d2c502ad6faad6818f853dadd1f8b8b182",
     "pandas>=2.0",
     "pyarrow>=14.0",
     "pyyaml>=6.0",

requirements.lock

@@ -66,11 +66,11 @@ pyyaml==6.0.3
     # via
     #   quant-data-kit
     #   quant-portfolio (pyproject.toml)
-quant-data-kit @ git+https://github.com/PureSaber/quant-data-kit.git@ba136c2fa2eea121bfb2ad7887b536c3952586f7
+quant-data-kit @ git+https://github.com/PureSaber/quant-data-kit.git@f85128307dd0e3407d24350282fe7b679b591309
     # via
     #   quant-execution
     #   quant-portfolio (pyproject.toml)
-quant-execution @ git+https://github.com/PureSaber/quant-execution.git@21aace45d2dde6458db8fcda4829f252f8c640f1
+quant-execution @ git+https://github.com/PureSaber/quant-execution.git@49ef53d2c502ad6faad6818f853dadd1f8b8b182
     # via quant-portfolio (pyproject.toml)
 referencing==0.37.0
     # via