Two bodies of work that touch the same files, so one commit.
Direction and setup-size campaign (2026-09-30 → 10-01, owner's RISK-FIRST rule,
written before the runs):
- SIGNAL: M15 daily-trend veto (factory override 7, xau_strategy.daily_trend_veto
"both"). A finished setup against a clear daily trend (closed D1 bars, Sunday
stubs dropped; close vs its 50-bar EMA AND vs the close 60 kept days earlier)
is removed as the last step of the core. One definition
(signal_aggregator.daily_trend_series) for live, the bar replay and the twin's
host side; D1 is a required frame of the M15 scan; mirrored in
cuda_signal_kernels.py K2. W3 824 / +205.12R / PF 1.73 / max DD 12.4R vs
1105 / +195.79R / 1.49 / 20.9R; 2019 → 2026 1473 / +223.33R / 1.42 / 41.5R vs
2091 / +192.53R / 1.24 / 64.8R; unseen holdout −4.62R vs −6.63R.
- FAILED, not shipped: the veto on M5 (W3 drawdown 10.1R → 15.2R) and on H1
(−9.7R), buy-only and sell-only on M15, the wide-stop cap
(xau_strategy.wide_stop_cap_atr stays 0.0, a research knob).
- New M15 pins: W2 dfc647ee…, OOS e87b6de1…, W3 ad3b0dbd…. M5 / H1 / H4 pins
unchanged. --set xau_strategy.daily_trend_veto=off reproduces the previous M15
pins byte for byte. Twin and reference script MD5-equal on every window;
recorder and feature-export parity re-proved; base-rate table rebuilt.
- DISPLAY (books byte-identical): exposure.py — the lot size for
trade_size.risk_per_trade (new config.json block, off by default, private to
the owner) and the count of alerted same-direction plans still open
(signals.alerted_at, read-only open_alerted_setups); the daily trend in words
on Telegram, the CLI and the dashboard; signals.daily_trend logged.
- providers/oanda: a date window with no bars is an empty page, not an error
(a D1 pad starting on a Saturday could not be fetched).
- Reports only, nothing shipped: seven context screens (no state nominated) and
eight outside data sources (none adds a forecast to gold's own daily trend).
- Independent verification (report section 5.13): every figure reproduced on
both scripts, scoring identical to the previous commit on all four
timeframes, no look-ahead. Caveats recorded there: this is a risk filter, not
a more accurate direction call; nine tenths of the 2019 → 2026 gain is 2022;
the neighbour clause is thin (a simple 100-day average misses the R line by
1.2R); "never adds a trade" is an observation, not a guarantee.
Bug review of the whole codebase (2026-10-01). Signal path clean. 17 defects
elsewhere, plus 4 found by an independent review of the fixes; all fixed, each
with a test; no trade book changed:
- telegram_bot: one _deliver_alert for the candle scan and the level break. The
row is stamped alerted_at BEFORE the send, on a one-worker alert-db pool, and
un-stamped when no chat received it or the task is cancelled mid-stamp, so
two alerts at one candle close count each other and no SQLite write runs on
the event loop. The 5-bar cooldown counts bars of the scan's own frame
(== gpu_backtest._apply_cooldown; clock time released the 22:00 UTC bar after
the settlement pause) and is restored from signals.db at start-up. The
lot-size line goes to the owner's own chat only (a group the owner ran
/autoscan in used to receive it). /status gains a news-model line and escapes
failure details.
- signal_log: mark_alerted returns the stamp it wrote; unmark_alerted;
last_alerted_bars (read-only).
- advisory_outcome: candle continuity is judged bar by bar inside the walk; a
hole after a trade's exit no longer voids it. The stored-candle replay
resolves 108 of 111 logged setups instead of 99, all equal to the live
resolver's R.
- providers/oanda: an empty date window is accepted only when it is 5 days or
shorter and no malformed candle was skipped; a longer one is retried, then
refused, instead of being cached as history with a hole.
- news_engine: blank RSS titles skipped; one unreadable headline no longer
costs the local model its whole batch (checked on the real bundle); model
health recorded; no DeepSeek audit call in a process that does not archive;
event-direction cache keyed per release; peek_week_events never fetches.
- llm_client: chat_json never raises and never logs the API key (requests
quotes the Authorization value in InvalidHeader; error bodies are redacted).
- signal_aggregator: the scan's news block is guarded; the replay path
(use_news=False) assigns exactly the values it did before.
- agent_panel: PANEL_PROMPT_VERSION v2.2 — same prompt text, corrected input
(event timing over the cached calendar week, headlines newest first).
scripts/panel_replay.py replays the logged payload per persona;
scripts/label_scan_outcomes.py counts with an ISO cutoff.
- test_all.py: no ALTER TABLE … DROP COLUMN (the live box has SQLite 3.34);
tests that write Parquet print a notice and return without pyarrow.
Verification after the last code change: suite 632 passed, 0 failed (also with
both Parquet engines blocked and FB2_BACKEND=cpu, the live box's view); the
five reference MD5s reproduce on both backtesters, 0 cache fallbacks; Module 37
green; static Python 3.9 / SQLite 3.34 check clean on every touched file.
Evidence: backtest/research/daily_trend_campaign_2026-09-30.md,
context_screens_2026-10.md, new_inputs_2026-10.md; CLAUDE.md sections 2, 4, 6,
8, 9, 10 and 11; pair_signals/xauusd.py docstring (M15 override 7).
Live signals.db review (2026-09-04 → 09-25, 100 unique setups replayed on the
scans' stored candles): the stop-hunt claim is refuted (71% of stop-outs ran
≥ 1R past the stop, the same as the books' 70–72%); −25.1R of the −37.8R raw
live book were consecutive-bar re-entries the backtester's 5-bar cooldown never
takes; the cooldown view sits at the 3rd–5th percentile of the books' three-week
windows.
Delivery (book-neutral):
- telegram_bot: 5-bar alert cooldown per timeframe (_ALERT_COOLDOWN_BARS =
backtest.cooldown_bars); setup block prints the trail first as the tested
exit, TP1/TP2 as guides with their R, "Planned TP1 RR", and one validated-book
line per setup class (agent_panel_base_rates.json); the informational 100-pip
TP line is removed (owner request).
- AI block rewritten for a new trader: "AI check: the odds look better than /
worse than / about usual", the chance "in 100" against the usual odds, one
reason for and one against, no verdict word or confidence figure; rendered
whenever the verdict carries a probability. agent_panel: persona prompts ban
trading jargon (PANEL_PROMPT_VERSION v2.1-2026-09-26); the macro-source
verdict now carries its persona's p_win (every live v2 row had
panel_probability NULL). Dashboard and guides mirror the wording.
- telegram_bot: level-breach rescans skip the settlement-pause window (the
20:45 M15 bar was re-logged at 22:01/22:06/22:11 every evening).
Signal (pre-registered rule-1 campaign, CUDA twin paired with gpu_backtest.py):
- M15 grade-A SELLs blocked (sell_min_quality ["A+","B"], M15 override 6):
W2 813/+154.15R PF 1.53 vs 931/+146.14R PF 1.43, OOS +68.5 vs +60.67,
W3 max DD 20.9 vs 19.3R. PASSED a–f.
- Printed target 1 floored at 1R in every generator (signals.tp_print_floor_rr),
applied AFTER the min-RR and spread gates so no trade changes; mirrored in
cuda_signal_kernels.py K2. Label-only re-pin of all four timeframes (only
tp1/rr_planned columns move). A floor applied before the gate cost −24.6R and
was rejected.
- FAILED: M15 min_rr_ratio 1.0 (trade-identical, −2.0R). Owner's call: M5
tp_floor_rr 1.0 (+48.9R W2, max DD 10.1 → 15.9R), not shipped.
New pins: M5 W2 e599c8a8…, M15 W2 5b65ecf6…, M15 OOS 09b8a8e7…, H1 W3
a44fd19b…, H4 W3 f3587bae…; twin and reference MD5s equal on every window.
Base-rate table rebuilt from the new pins. Suite 558 passed.
Evidence: backtest/research/rr_and_sell_campaign_2026-09-26.md (plus the
2026-09-18 loss-anatomy and wrong-direction reports it builds on).
Owner-approved 2026-09-09 after the pre-registered weight test
(backtest/research/weight_test_2026-09-09.md). M15-only
technical.regime_weights_override = the shipped table with rsi x 0.5
(0.5 trending family / 0.75 ranging / 0.5 transitioning), derived in the
factory and pinned literally by the Module 9 equivalence test; a Module 33
test locks that M5/H1/H4 carry no override.
Evidence, 2-year Oanda window 2024-09-05..2026-09-05 at $0.65: 793/+117.62R
PF 1.41 -> 931/+146.14R PF 1.43 (fixed halves +27.9/+0.6R vs control); OOS
2024-06-01..2025-03-30 300/+40.59R PF 1.38 -> 355/+60.67R PF 1.49; 3-year
+151.90R -> +186.61R (max DD 15.5 -> 19.3R, 2026 -4R). Parity: the factory
run with no --set flag reproduces the validated candidate books byte-for-
byte (2-yr MD5 d185f099..., OOS 6f8f24b4...). M15 canonical book re-pinned
to backtest/oanda_m15_2yr_065_20260909.csv. Docs carry a one-line note.
Claude-Session: https://claude.ai/code/session_01Nc9Ninzyi18QkWezrxeBvB
2026-08-04 audit, Batch D. Verification outcomes (no behavior changes —
comment + tests only):
- CP gate-bypass (quality forced to B after the gate chain, exhaustion/
M5-score gates not re-checked) is VALIDATED behavior, not a bug: the
pinned shipped-config M5 12mo book carries 226/586 XAU_CP trades
(+38.9R), 104 with EXHAUST warnings (+15.3R), 175 below the M5
min-score gate (+28.0R). Removing the "holes" removes validated
profit. New test pins the shipped state (PB off, CP BUY-only on M5,
ema_partial_neutral M5-only) so any flip is a conscious revalidated
decision.
- E18 (xau_strategy vs backtest blocked-hours "drift") REFUTED as a
defect: the divergence is required by pair isolation — the backtest
copy is the generic non-XAU fallback and must not inherit
gold-validated deltas. Documented in the factory + a test asserting
the delta stays XAU-side.
- 6c-upgrade re-arming an exhaustion-D refuted (exhaustion zeroes
confidence below 6c's floor; regimes mutually exclusive). "WEEKEND
FILTER" CP-suppression branch is defense-in-depth for direct
analyze_from_dataframes callers, kept.
Suite: 316 green.
Claude-Session: https://claude.ai/code/session_01UnZkzeFcTu9agFg8BDh2tn
Gate 6h (big-candle entry gate): blocks all XAU entries (core, XAU_PB,
XAU_CP) when the primary TF's last 24h contains a bar with range >= K x
prior-bar ATR. Ships default-OFF: validation rejected every threshold
(K 3-6 on M5/M15 12mo) — removed trades were net-POSITIVE at every cell;
M5 quiet-window trades are the only losing bucket. Config knobs
big_candle_atr_multiple/lookback_hours, --big-candle backtest flag,
Module 17 tests.
M5 relaxation SHIPPED (validated 2026-07-30, Oanda bars, $0.30 spread):
m5_min_abs_score 30->25, m5_min_confirmation_score 10->5, hour-21 UTC
unblocked M5-only. Beats control in ALL 6 calendar years of the 5yr
replay (+230.9R/2319 PF 1.26 DD 21.3 vs +168.0R/2048 PF 1.21 DD 27.0),
12mo +107.4R PF 1.36 vs +84.6R PF 1.33, OOS +26.1R vs +17.2R, $0.70
spread +86.9R vs +66.4R. M15: every relaxation retested negative —
config unchanged. Declined-but-open: M5 hour-22 block, M15 hour-16
unblock (both fail 5yr/OOS). Evidence in the factory docstring.
Tooling: generic --set PATH=JSON research override in gpu_backtest
(typo-proof, logged); diagnose_filters filter_tags refreshed for gates
6d-6i with a staleness-guard test (Module 18); research scripts
hour_analysis/burst_profile/book_diff. Suite: 268 tests.
Claude-Session: https://claude.ai/code/session_014jqHpmNYmwYMowF6bPZyeD
Same-day sibling of the M15 campaign, opposite finding: on M5 the arm
point barely matters — winners need a looser leash. 5yr grid plateau
arm 0.5-0.7 x dist 0.85-1.0; shipped the OOS-preferred edge 0.60/0.85:
+231.2R PF 1.30 vs +218.0R PF 1.29 baseline, both halves better, DD
flat (19.2 vs 19.1), OOS +14.5R vs +11.1R. dist 1.0 rejected at the
OOS gate despite best in-sample total. XAU_PB pb_trail unaffected.
Factory replay reproduces the grid cell; 201/201 tests.
Claude-Session: https://claude.ai/code/session_01VPNF37MP9Kuv62B1VBrZ1b
Parity: replay now applies the live _apply_regime_filter_live method
under the live guard (function-identity, default-off); unregistered-pair
fallback applies the config.json overlay; parity sweep gains direct
ATR/vol-regime assertions; regime-filter semantics test added.
Trail campaign: --trail-arm/--trail-dist research flags; 30-cell grid on
5yr M15 refuted the earlier-arm hypothesis and found the 0.70-0.80 arm
plateau. Shipped M15-only trail_arm_r=0.70 (factory override):
+210.7R PF 1.35 WR 57.8% vs +175.7R PF 1.30 WR 53.9% baseline; both
halves and 2024-06->2025-03 OOS better (+57.4 vs +45.4); 5yr max-DD
clause (33.9R vs 29.1R) explicitly waived by owner. M5 unchanged.
Factory replay reproduces the grid cell trade-for-trade; 201/201 tests.
Claude-Session: https://claude.ai/code/session_01VPNF37MP9Kuv62B1VBrZ1b
Implements Task 2: apply_pair_overrides in xauusd.make_config before
the validated per-TF deltas, so e.g. M5 hour-9 appends on top of the
JSON base. Updated Module 9 equivalence test and added three parity pins.
Claude-Session: https://claude.ai/code/session_01VPNF37MP9Kuv62B1VBrZ1b
M5 factory: cp_enabled=True, patterns limited to inv_hs, double_bottom,
bull_flag. SELL CP lost -3.21R on OOS; BUY CP: +41.8R PF 1.60 on 12mo,
+65.2R PF 1.48 on 5yr. Adds +22R to the M5 book (708 vs 513 trades).
M15/H1 remain disabled — all failed OOS validation.
Claude-Session: https://claude.ai/code/session_015kby7M9JwpYi8NQT2yoxAi
M15 mixed-"trending" regime block (ADX strong, EMAs unaligned =
momentum without direction): -6.8R/24 trades on 12mo, negative both
halves; blocking scores +65.7R/499 PF 1.38 DD 10.4R vs +59.0R/521
PF 1.32 DD 12.6R, better both halves, OOS 2024-06->2025-03 neutral
(-1.4R = the 4 such trades there). M5 never trades this regime.
New XauStrategyConfig.blocked_regimes / sell_blocked_regimes knobs
(default empty) + aggregator gate 6g2; M15 factory opts in.
M5 hour-9 UTC block (post-London-open whipsaw): -28.9R/143 over the
6-year book, negative 5/7 calendar years. Full-history replay with
block: +207.1R vs +178.3R, max DD 28.7R vs 35.5R, 2023 -13.0R ->
breakeven; tuned 12mo book unchanged (139.0 vs 139.3). Hour 9 is
positive on M15 (+1.7R 6yr, +8.6R OOS) so the block is factory
M5-only.
Rejected after validation (documented in xauusd.py + session memory):
transitioning-SELL block (bull-market artifact, +5.7R in 2024 OOS),
M15 trending_down block (M5 cross-TF contradiction), losing-streak
circuit breakers, 30d efficiency-ratio trend gate, and all entry-level
indicator filters (winners == losers at entry on every feature).
Full-history context (Twelve Data 5min starts 2020-03-16): M5 6.3yr
+207R PF ~1.3 post-fix with losing year 2021; M15 +152.6R PF 1.19
DD 49.7R — edge is regime-dependent, chop years are structural.
Tests: 178 passed (new gate test + factory-equivalence updates).
Finishes PENDING_STEPS Step 1 under the current trailing-exit model.
M15 window 2025-03-30->2026-05-31 ($0.30 spread): +116.6R/460 PF 1.83
vs +105.4R/424 PF 1.82 baseline, better in both halves; 2024-06->
2025-03 OOS +30.3R vs +10.6R control. Same knob on the pinned M5 book
is WORSE (+131.3R/699 vs +139.9R/547) so the override is M15-only in
the xauusd factory, per the pre-registered decision rule. M5 path
unchanged (factory override gated on primary_tf == M15).
Ported from feature/eurusd-support: PAIR_PRODUCERS registry with
make_config() factories per pair. xauusd.make_config is a pure
PlatformConfig-defaults wrapper (equality with the inline construction
is enforced by test); eurusd.make_config carries the donor branch's
regime weights and gates as a pre-realism baseline.
Adds 4 registry tests. 154/154 pass.
Harwinder Sekhon