Harwinder Sekhon b8f903ad0e agent_panel: alerts use the macro persona's verdict; all 4 takes still logged ...

agent_panel.verdict_source defaults to "macro": the macro take's stance is the
verdict, the synthesizer call is skipped (4 LLM calls per panel instead of 5);
"chair" restores the four-voice consensus + OPPOSE guard. Verdict source is
cached per mode and logged (panel_verdict_source). 406 tests.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FC4s6VCURFte4dub4xKxYL

2026-09-03 21:16:45 -04:00

backtest

xauusd factory: block M5 shorts in transitioning regime, block all H1 shorts (rule-1 validated)

2026-09-03 20:47:33 -04:00

deploy

dashboard: clickable news — detail dialogs + on-demand AI summary endpoint

2026-08-07 09:08:49 -04:00

docs

Remove Oanda auto-execution daemon/executor (not auto-trading)

2026-09-03 21:00:21 -04:00

pair_signals

xauusd factory: block M5 shorts in transitioning regime, block all H1 shorts (rule-1 validated)

2026-09-03 20:47:33 -04:00

providers

signal_log: persist agent panel verdict + per-persona takes to SQLite

2026-08-14 20:41:54 -04:00

scripts

signal_log: outcomes for every setup, event dedup, interned news; fix over-scan, dead outlook, panel OPPOSE bias

2026-09-03 20:47:33 -04:00

static

dashboard: redesign signal workspace and refresh flow

2026-08-18 14:54:47 -04:00

.gitignore

news: DeepSeek validation, vs_forecast fix, /status source health

2026-08-06 22:21:02 -04:00

advisory_outcome.py

agent_panel: harden evidence and track outcomes

2026-08-23 12:57:05 -04:00

agent_panel.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

analyze_signal_log.py

signal_log: outcomes for every setup, event dedup, interned news; fix over-scan, dead outlook, panel OPPOSE bias

2026-09-03 20:47:33 -04:00

app_config.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

chart_patterns.py

Accuracy audit fixes: alerts carry validated per-TF trail plans + 13 more

2026-07-23 10:31:13 -04:00

CLAUDE.md

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

config.json.example

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

config.py

signal_log: persist agent panel verdict + per-persona takes to SQLite

2026-08-14 20:41:54 -04:00

dashboard_publisher.py

Remove Oanda auto-execution daemon/executor (not auto-trading)

2026-09-03 21:00:21 -04:00

engine.py

M5 at real cost: disable XAU_PB, squash partial EMA alignment

2026-08-04 16:54:09 -04:00

llm_client.py

news: DeepSeek validation, vs_forecast fix, /status source health

2026-08-06 22:21:02 -04:00

llm_provider_switch_plan.txt

docs: multi-provider LLM layer plan (DeepSeek news / Anthropic panel)

2026-08-07 23:02:08 -04:00

main.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

market_outlook.py

signal_log: outcomes for every setup, event dedup, interned news; fix over-scan, dead outlook, panel OPPOSE bias

2026-09-03 20:47:33 -04:00

news_engine.py

news: reduce headline max age from 24h to 12h, make configurable

2026-08-09 16:49:48 -04:00

requirements.txt

Refocus on XAU/USD: add momentum indicators, tune TP, exhaustion block

2026-05-15 10:51:51 -04:00

signal_aggregator.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

signal_log.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

telegram_bot.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

test_all.py

agent_panel: alerts use the macro persona's verdict; all 4 takes still logged

2026-09-03 21:16:45 -04:00

S

Description

No description provided

28 MiB

Languages

Python 92.4%

HTML 6.6%

PHP 1%