Clarify IPO short-exit strategy horizon

Request:
- Emphasize that the analyst model is for selling allocated IPO shares in T2 grey market or on D1, not for long-term holding.

Changes:
- Add explicit T2/D1 sell discipline to the analyst skill.
- Update ipo_score_v0 targets and holding policy so D1 sell return is primary and T2 is the intended extension when reliable grey-market data exists.
- Clarify that D5/D20/D60 are review labels only, not planned holding targets.
- Update the model report, single-ticker report generator, README, and the 06106 report language to reflect the short-exit horizon.

Verification:
- Rebuilt the model report with the same dataset timestamp and confirmed the analysis dataset did not change.
- Ran py_compile for build_analysis_dataset.py and generate_ipo_report.py.
- Generated a 06106 dry-run report showing T2/D1 exit discipline.
- Ran git diff --check.

Next useful context:
- T2 is still disabled in v0 until archivist approves a reliable grey-market data source; D1 remains the live modeled sell label.

.codex/skills/analyst/SKILL.md

@@ -7,7 +7,7 @@ description: Use for Hong Kong IPO subscription analysis in this project: T0/T1/
 
 ## Purpose
 
-Assess Hong Kong IPO subscription candidates using the project's archived facts, scoring rules, prediction cards, and review history. This skill owns judgment: whether to subscribe, wait, avoid, hold after allocation, or revise a rule after outcomes arrive.
+Assess Hong Kong IPO subscription candidates using the project's archived facts, scoring rules, prediction cards, and review history. This skill owns judgment: whether to subscribe, wait, avoid, sell in grey market or on D1, or revise a rule after outcomes arrive.
 
 Use `archivist` first when source documents, listing facts, allotment results, prices, or database snapshots need to be updated.
 
@@ -22,6 +22,16 @@ Separate the decision stage from later facts:
 
 Do not let later facts leak into earlier prediction cards. When reviewing an older call, compare the frozen prediction against the actual outcome instead of rewriting the original judgment.
 
+## Trading Horizon
+
+The analyst model is a short-exit IPO subscription model, not a long-term holding model.
+
+- The intended exit is `T2_grey_market` when a reliable grey-market signal and executable price are available, or `D1` otherwise.
+- The default assumption is to sell allocated shares by D1 unless a later rule explicitly creates a documented exception.
+- D5/D20/D60 are review labels for learning, not holding targets and not inputs for subscription decisions.
+- Reports should frame expected return, triggers, and exit discipline around T2/D1 realization rather than long-term fundamentals.
+- Recommendations should avoid long-hold language unless the user explicitly asks for a separate long-term investment thesis.
+
 ## Project Storage Contract
 
 Use repo-relative paths everywhere:
@@ -77,7 +87,7 @@ Generated prediction reports must remain stage-safe:
 
 - T0 reports use only prospectus-stage fields and T0 calibration.
 - T1 reports may add allotment demand fields and T1 calibration.
-- D1/D5/D20/D60 returns are never shown as prediction inputs; they are reserved for later review cards.
+- T2/D1 is the intended sell window; D5/D20/D60 returns are never shown as prediction inputs and are reserved for later review cards.
 
 ## Output Standards
 
@@ -96,6 +106,7 @@ Every prediction card should include:
 - key risks
 - triggers for upgrade/downgrade
 - exit plan
+- explicit T2/D1 sell discipline
 - source paths
 
 Every review card should include:

README.md

@@ -9,6 +9,8 @@ The project is designed around a feedback loop:
 3. Compare predictions with post-listing outcomes.
 4. Improve the scoring rules only from reviewed evidence.
 
+The investment horizon is deliberately short. The subscription model is built for selling allocated shares in `T2_grey_market` when a reliable executable grey-market signal exists, or on `D1` otherwise. It is not a long-term holding model.
+
 ## Goals
 
 - Maintain a local, Git-tracked history of Hong Kong IPO data.
@@ -28,6 +30,8 @@ Each IPO is evaluated by stage:
 
 The key discipline is to avoid hindsight leakage. A T0 prediction should only use T0 information, even after the IPO has listed.
 
+The default exit discipline is T2/D1. D5/D20/D60 outcomes are kept to review whether the model missed later information, not to justify holding IPO allocations beyond the intended sell window.
+
 ## Project Skills
 
 This repository includes project-local Codex skills under `.codex/skills/`.
@@ -153,7 +157,7 @@ Use the analyst model builder to digest archived data into a stage-safe scoring
 
 The v0 model is documented in `rules/ipo_score_v0.yaml`. It writes `data/snapshots/analysis_model_v0_dataset.csv` and `reports/2026-06-15_analysis_model_v0.md`.
 
-The model separates T0 prospectus inputs from T1 allotment inputs. D1/D5/D20/D60 returns are labels for calibration and review, not prediction inputs.
+The model separates T0 prospectus inputs from T1 allotment inputs. Its live trading target is a T2 or D1 sale. D5/D20/D60 returns are labels for calibration and review, not prediction inputs or planned holding targets.
 
 ## Single IPO Markdown Report
 
@@ -166,7 +170,7 @@ Use the analyst report generator after the archive and model dataset are current
 
 The generator writes `reports/{date}_{ticker}_{stage}_analysis.md` by default. It auto-selects `T1_allotment` when structured allotment-demand facts exist; otherwise it generates a `T0_prospectus` report. Use `--stdout` for a dry run or `--output` to choose a specific Markdown path.
 
-Prediction reports are stage-safe: T0 reports use only prospectus-stage facts and T0 calibration, while T1 reports add allotment demand and T1 calibration. Post-listing D1/D5/D20/D60 performance stays out of prediction reports and is reserved for review cards.
+Prediction reports are stage-safe: T0 reports use only prospectus-stage facts and T0 calibration, while T1 reports add allotment demand and T1 calibration. Reports should frame the trade as a T2/D1 exit. Post-listing D5/D20/D60 performance stays out of prediction reports and is reserved for review cards.
 
 ## Incremental Archive Sync
 

reports/2026-06-15_06106_T0_prospectus_analysis.md

@@ -9,8 +9,9 @@
 - Model dataset as of: `2026-06-15T14:04:34Z`
 - Rule version: `ipo_score_v0`
 - Rule path: `rules/ipo_score_v0.yaml`
+- Strategy horizon: short IPO subscription trade; intended exit is T2 grey market if reliable, otherwise D1.
 - Decision: `strong_watch`
-- PM action: Strong watch at T0, still pending T1 demand confirmation.
+- PM action: Strong watch at T0, still pending T1 demand confirmation for a T2/D1 exit.
 - T0 score: `11`
 - Score bucket: `t0_gte_8`
 - Calibrated D1 positive probability: 76.4% from 72 historical D1 labels
@@ -35,12 +36,14 @@
 | Initial public offer percentage | 5.0% |
 | Over-allotment shares | 1,574,550 |
 
-## Model Inference
+## Short-Exit Model Inference
 
 - D1 positive probability: 76.4%
 - D1 >= 10% probability: 47.2%
 - Historical average D1 return for bucket: 28.6%
 - Historical median D1 return for bucket: 9.6%
+- T2 sell return is not modeled until an approved grey-market data source exists.
+- D5/D20/D60 outcomes are review labels only, not holding targets.
 
 ## Score Breakdown
 
@@ -64,8 +67,8 @@
 
 - No material negative scoring component.
 - No required report field is blank for this stage.
-- T2 grey-market signal is not used because the project has no approved reproducible source.
+- T2 grey-market signal is not used yet because the project has no approved reproducible source.
-- Post-listing D1/D5/D20/D60 outcomes are labels for model calibration only and are not shown as prediction inputs.
+- Post-listing D5/D20/D60 outcomes are labels for later review only and are not holding-period targets.
 
 ## Triggers
 
@@ -74,9 +77,10 @@
 
 ## Exit Plan
 
-- If subscribed and allocated, reassess after allotment and before first trading session using only information available at that stage.
+- If subscribed and allocated, plan to sell in T2 grey market when reliable executable data is available.
-- For T1 reports without approved T2 data, treat first-day liquidity and position sizing conservatively.
+- If T2 is unavailable or unreliable, use D1 as the default exit window.
-- Record actual D1/D5/D20/D60 outcomes later as review labels, not as retroactive prediction inputs.
+- Do not treat D5/D20/D60 as planned holding periods for this model.
+- Record D1/D5/D20/D60 outcomes later as review labels, not as retroactive prediction inputs.
 
 ## Source Paths
 

reports/2026-06-15_analysis_model_v0.md

@@ -7,9 +7,9 @@
 
 ## What This Model Does
 
-This is the first analyst model built from the downloaded archive. It creates a repeatable feature table, scores each IPO using stage-safe rules, and calibrates the score buckets against archived D1 outcomes. It is intentionally transparent: the output includes every score component and the archived source paths used for each ticker.
+This is the first analyst model built from the downloaded archive. It creates a repeatable feature table, scores each IPO using stage-safe rules, and calibrates the score buckets against archived D1 sell outcomes. It is intentionally transparent: the output includes every score component and the archived source paths used for each ticker.
 
-The model does not use grey-market data in v0 because T2 currently has no approved reproducible source. It also does not use post-listing returns as inputs; returns are labels only.
+The model is built for a short IPO allocation trade: sell in T2 grey market when reliable executable data exists, or sell on D1 otherwise. It does not use grey-market data in v0 because T2 currently has no approved reproducible source. It also does not use post-listing returns as inputs; D1 is the primary sell label, while D5/D20/D60 are review labels only.
 
 ## Data Inventory
 
@@ -58,7 +58,8 @@ The high-conviction bucket remains clearly differentiated, but the middle and lo
 1. Run `scripts/build_analysis_dataset.py` after archivist updates the database.
 2. Use `t0_score` for prospectus-stage watchlisting.
 3. Use `total_score`, `decision_band`, and `calibrated_d1_positive_rate` for T1-stage subscription cards.
-4. Treat D1/D5/D20/D60 columns as review labels only, never as prediction inputs.
+4. Frame live decisions around a T2 or D1 sell, not long-term holding.
+5. Treat D5/D20/D60 columns as review labels only, never as prediction inputs or holding targets.
 
 ## Known Gaps
 

rules/ipo_score_v0.yaml

@@ -7,14 +7,27 @@ purpose: >
   Build the first transparent Hong Kong IPO subscription scoring model from
   archived project data. The model is a rules-plus-calibration baseline, not a
   black-box predictor. It separates T0 prospectus information from T1 allotment
-  information so later facts do not leak into earlier decisions.
+  information so later facts do not leak into earlier decisions. It is designed
+  for short IPO subscription exits, not long-term holding.
 
 targets:
-  primary: D1 return versus IPO offer price
+  primary: D1 sell return versus IPO offer price
   secondary:
-    - D5 return versus IPO offer price
+    - T2 grey-market sell return versus IPO offer price, disabled until a reliable source exists
-    - D20 return versus IPO offer price
+    - D1 strong-return probability
-    - D60 return versus IPO offer price
+
+holding_policy:
+  intended_exit_window:
+    - T2_grey_market when a reliable grey-market price is available
+    - D1 otherwise
+  default_assumption: sell allocated shares by D1
+  non_targets:
+    - D5 return
+    - D20 return
+    - D60 return
+  note: >
+    D5/D20/D60 outcomes are retained for review and error attribution only.
+    They do not define the subscription model's target holding period.
 
 stage_policy:
   T0_prospectus:
@@ -147,4 +160,4 @@ known_limitations:
   - T1 is structurally complete for listed rows, but some individual demand fields remain null when the archived source does not explicitly state them.
   - Industry labels and issuer fundamentals are not yet sufficiently structured for v0.
   - T2 grey-market signal is intentionally disabled until a reliable source exists.
-  - Current D1 outcomes are from archived market data and should be reviewed for extreme corporate-action or data-vendor anomalies.
+  - Current D1 sell-return outcomes are from archived market data and should be reviewed for extreme corporate-action or data-vendor anomalies.

rules/rule_change_log.md

@@ -1,5 +1,29 @@
 # Rule Change Log
 
+## 2026-06-15 - Clarify short-exit IPO strategy horizon
+
+Request:
+
+- Emphasize that the analyst model is focused on selling allocated IPO shares in T2 grey market or on D1, not long-term holding.
+
+Change:
+
+- Added an explicit T2/D1 sell horizon to the analyst skill.
+- Updated `ipo_score_v0` targets and holding policy to make D1 sell return the primary modeled label and T2 the intended extension when reliable grey-market data exists.
+- Clarified that D5/D20/D60 are review labels only, not holding-period targets.
+- Updated single-ticker reports and the report generator to show T2/D1 exit discipline.
+
+Rationale:
+
+- The subscription decision should optimize for immediate IPO exit execution, not a long-term equity thesis.
+- This preserves stage safety while aligning report language, model targets, and review labels with the actual trading process.
+
+Verification:
+
+- Generated dry-run single-ticker reports after the template update.
+- Rebuilt the analysis model report with the same dataset timestamp to refresh language without changing model rows.
+- Ran py_compile for the modified scripts and checked Markdown/report text for the new T2/D1 discipline.
+
 ## 2026-06-15 - Refresh `ipo_score_v0` after T1 demand backfill
 
 Request:

scripts/build_analysis_dataset.py

@@ -611,9 +611,9 @@ def write_report(
         "",
         "## What This Model Does",
         "",
-        "This is the first analyst model built from the downloaded archive. It creates a repeatable feature table, scores each IPO using stage-safe rules, and calibrates the score buckets against archived D1 outcomes. It is intentionally transparent: the output includes every score component and the archived source paths used for each ticker.",
+        "This is the first analyst model built from the downloaded archive. It creates a repeatable feature table, scores each IPO using stage-safe rules, and calibrates the score buckets against archived D1 sell outcomes. It is intentionally transparent: the output includes every score component and the archived source paths used for each ticker.",
         "",
-        "The model does not use grey-market data in v0 because T2 currently has no approved reproducible source. It also does not use post-listing returns as inputs; returns are labels only.",
+        "The model is built for a short IPO allocation trade: sell in T2 grey market when reliable executable data exists, or sell on D1 otherwise. It does not use grey-market data in v0 because T2 currently has no approved reproducible source. It also does not use post-listing returns as inputs; D1 is the primary sell label, while D5/D20/D60 are review labels only.",
         "",
         "## Data Inventory",
         "",
@@ -652,7 +652,8 @@ def write_report(
         "1. Run `scripts/build_analysis_dataset.py` after archivist updates the database.",
         "2. Use `t0_score` for prospectus-stage watchlisting.",
         "3. Use `total_score`, `decision_band`, and `calibrated_d1_positive_rate` for T1-stage subscription cards.",
-        "4. Treat D1/D5/D20/D60 columns as review labels only, never as prediction inputs.",
+        "4. Frame live decisions around a T2 or D1 sell, not long-term holding.",
+        "5. Treat D5/D20/D60 columns as review labels only, never as prediction inputs or holding targets.",
         "",
         "## Known Gaps",
         "",

scripts/generate_ipo_report.py

@@ -216,13 +216,13 @@ def action_for_decision(decision: str) -> str:
     actions = {
         "weak_or_avoid": "Avoid at T0 unless later T1 demand changes the setup.",
         "neutral": "Wait for T1 allotment demand before subscribing.",
-        "positive_watch": "Watch positively, but wait for T1 confirmation before sizing.",
+        "positive_watch": "Watch positively, but wait for T1 confirmation before sizing for a T2/D1 exit.",
-        "strong_watch": "Strong watch at T0, still pending T1 demand confirmation.",
+        "strong_watch": "Strong watch at T0, still pending T1 demand confirmation for a T2/D1 exit.",
         "avoid": "Avoid subscription.",
         "avoid_or_wait": "Avoid or wait; do not size without a stronger catalyst.",
-        "watch_or_small": "Small subscription only if execution constraints are favorable.",
+        "watch_or_small": "Small subscription only if execution constraints support a T2/D1 exit.",
-        "selective_subscribe": "Selective subscription with disciplined sizing.",
+        "selective_subscribe": "Selective subscription with disciplined T2/D1 sell sizing.",
-        "high_conviction_subscribe": "Subscribe, subject to allocation and liquidity discipline.",
+        "high_conviction_subscribe": "Subscribe, subject to allocation, liquidity, and T2/D1 sell discipline.",
     }
     return actions[decision]
 
@@ -360,6 +360,7 @@ def build_report(record: dict[str, str], rows: list[dict[str, str]], stage: str,
         f"- Model dataset as of: `{dataset_as_of}`",
         f"- Rule version: `{model_version}`",
         f"- Rule path: `{MODEL_RULE_PATH.as_posix()}`",
+        "- Strategy horizon: short IPO subscription trade; intended exit is T2 grey market if reliable, otherwise D1.",
         f"- Decision: `{decision}`",
         f"- PM action: {action_for_decision(decision)}",
         f"- {score_label}: `{score}`",
@@ -370,12 +371,14 @@ def build_report(record: dict[str, str], rows: list[dict[str, str]], stage: str,
         "",
         facts_table(record, stage),
         "",
-        "## Model Inference",
+        "## Short-Exit Model Inference",
         "",
         f"- D1 positive probability: {fmt_pct_rate(metric.d1_positive_rate)}",
         f"- D1 >= 10% probability: {fmt_pct_rate(metric.d1_strong_rate)}",
         f"- Historical average D1 return for bucket: {fmt_num(metric.average_d1_return_pct, '%')}",
         f"- Historical median D1 return for bucket: {fmt_num(metric.median_d1_return_pct, '%')}",
+        "- T2 sell return is not modeled until an approved grey-market data source exists.",
+        "- D5/D20/D60 outcomes are review labels only, not holding targets.",
         "",
         "## Score Breakdown",
         "",
@@ -389,8 +392,8 @@ def build_report(record: dict[str, str], rows: list[dict[str, str]], stage: str,
         "",
         *reason_lines(components, positive=False),
         *missing_field_lines(record, stage),
-        "- T2 grey-market signal is not used because the project has no approved reproducible source.",
+        "- T2 grey-market signal is not used yet because the project has no approved reproducible source.",
-        "- Post-listing D1/D5/D20/D60 outcomes are labels for model calibration only and are not shown as prediction inputs.",
+        "- Post-listing D5/D20/D60 outcomes are labels for later review only and are not holding-period targets.",
         "",
         "## Triggers",
         "",
@@ -399,9 +402,10 @@ def build_report(record: dict[str, str], rows: list[dict[str, str]], stage: str,
         "",
         "## Exit Plan",
         "",
-        "- If subscribed and allocated, reassess after allotment and before first trading session using only information available at that stage.",
+        "- If subscribed and allocated, plan to sell in T2 grey market when reliable executable data is available.",
-        "- For T1 reports without approved T2 data, treat first-day liquidity and position sizing conservatively.",
+        "- If T2 is unavailable or unreliable, use D1 as the default exit window.",
-        "- Record actual D1/D5/D20/D60 outcomes later as review labels, not as retroactive prediction inputs.",
+        "- Do not treat D5/D20/D60 as planned holding periods for this model.",
+        "- Record D1/D5/D20/D60 outcomes later as review labels, not as retroactive prediction inputs.",
         "",
         "## Source Paths",
         "",