Pre-Trade Discipline Gate

Check a local pre-trade discipline checklist before manual broker entry, blocking planless, oversized, revenge-risk, market-regime-blocked, or circuit-breaker-blocked orders while journaling the result.

No API

View Source on GitHub

Table of Contents

1. Overview

Pre-Trade Discipline Gate runs after a candidate has been validated, sized, and journaled, but before any manual broker order is placed. It produces a pre_trade_discipline_decision artifact with a candidate-by-candidate checklist result.

The skill is offline and advisory. It does not place orders.

2. Quick Start 

python3 skills/pre-trade-discipline-gate/scripts/check_pre_trade_discipline.py \
  --answers-file state/manual-entry-checklist.json \
  --state-dir state/theses \
  --market-regime-decision reports/exposure_decision_latest.json \
  --circuit-breaker-decision reports/circuit_breaker_decision_latest.json \
  --output-dir reports/pre-trade-discipline \
  --journal-dir state/journal/pre-trade-discipline

3. Checklist Input 

{
  "candidates": [
    {
      "symbol": "AAPL",
      "thesis_id": "th_aapl_gm_20260703_0001",
      "order_intent": "ENTRY_READY",
      "entry_in_written_plan": true,
      "stop_predefined": true,
      "size_within_plan": true,
      "planned_risk_dollars": 500,
      "actual_risk_dollars": 500
    }
  ]
}

Actionable intents are ENTRY_READY, ACTIONABLE, ACTIONABLE_DAY1, and MANUAL_ORDER. Watchlist or ignored candidates are recorded as NO_ACTIONABLE_ORDERS.

4. Decisions

Decision Meaning
GO All actionable manual-order candidates passed
REVIEW_REQUIRED Missing, unknown, or failed journaling inputs need review
NO_GO A discipline or upstream gate rule blocked an actionable order
NO_ACTIONABLE_ORDERS The file contains no broker order to place

Use --fail-on-non-go when shell automation should return exit code 2 for non-GO decisions.

When a candidate has thesis_id and --state-dir is supplied, the generated JSON report is added to the thesis linked_reports list. The skill does not call mark_reviewed, so monitoring review dates are not advanced.

The JSON report and JSONL journal keep each candidate’s checklist_answers, including written-plan, stop, size, risk-dollar, and notes fields, so later reviews can audit what was answered before the order.