Atlas 技術版 > Prospective Shadow Exit Runner Intraday

Prospective Shadow Exit Runner Intraday NO_AGENT UNKNOWN risk=MEDIUM

1. 做什麼 / 2. 為什麼存在

(purpose 尚未填寫 — manifest 為 draft/INFERRED)

3. Schedule / Mode

mode=NO_AGENT state=scheduled enabled=1
schedule=*/10 9-12 * * 1-5
next_run=2026-10-21T09:00:00 last_run=2026-10-20T12:50:16
deliver=local model=— provider=—

4. Implementation

script=/Users/exion/.hermes/scripts/prospective_shadow_exit_runner.py
monitor_script=— monitor_url=—
skills=[]
workdir=—

5. Effective instruction sources

prompt_sha256=e8ed90c83981b156… chars=397
skill_files={}

Execute /Users/exion/.hermes/scripts/prospective_shadow_exit_runner.py. Load only OBSERVED Formal entries from the prospective shadow DB, run deterministic Stop/Stale/Profit state updates, and record simulated execution outcomes only. Use stop structural_baseline/price_only, profit ma20/atr_3, scale-out 33/33/34. Never place orders, send Telegram, change production config, or enable auto-order.…

6. Actual rules / thresholds

rule	value	source
none extracted

7/8. Inputs & Outputs (resources)

kind	path	name	access
SQLITE_DB	/Users/exion/.hermes/data/prospective_exit_shadow.db		unknown
SQLITE_TABLE	self	shadow_entry_snapshots	read
SQLITE_TABLE	self	discovery_intraday_bars	read
SQLITE_TABLE	self	discovery_intraday_bars	read
SQLITE_TABLE	self	stock_kline	read
SQLITE_TABLE	self	discovery_intraday_bars	read
SQLITE_TABLE	self	discovery_intraday_bars	read
SQLITE_TABLE	self	shadow_metadata	read
SQLITE_TABLE	self	shadow_metadata	read
SQLITE_TABLE	self	shadow_lifecycle_events	read
SQLITE_TABLE	self	shadow_entry_snapshots	read
SQLITE_TABLE	self	shadow_exit_events	read
SQLITE_TABLE	self	shadow_execution_outcomes	read
SQLITE_TABLE	self	stock_kline	read
SQLITE_TABLE	self	stock_kline	read
SQLITE_TABLE	self	discovery_exit_replay_runs	read
SQLITE_TABLE	self	sqlite_master	read
SQLITE_TABLE	self	shadow_metadata	write
SQLITE_TABLE	self	shadow_lifecycle_events	write
SQLITE_TABLE	self	shadow_entry_snapshots	write
SQLITE_TABLE	self	shadow_exit_events	write
SQLITE_TABLE	self	shadow_execution_outcomes	write
SQLITE_TABLE	self	discovery_exit_replay_runs	write
SQLITE_TABLE	self	discovery_trade_outcomes	write
SQLITE_TABLE	self	discovery_exit_variant_results	write
SQLITE_TABLE	self	discovery_exit_state_events	write
FILE_WRITE	+00:00		write
FILE_WRITE	+00:00		write
FILE_WRITE	T		write
FILE_WRITE	+00:00		write
MODULE	/Users/exion/workspace/investment_core/discovery_engine/shadow_exit_runner.py	investment_core.discovery_engine.shadow_exit_runner	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/config.py	config	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/exit_audit_v3.py	exit_audit_v3	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/profit_exit_engine.py	profit_exit_engine	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/stale_exit_engine.py	stale_exit_engine	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/stop_engine.py	stop_engine	import
MODULE	/Users/exion/workspace/investment_core/discovery_engine/transaction_replay.py	transaction_replay	import
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown
CONFIG		UNKNOWN	unknown

9/10. Upstream / Downstream

Upstream: Shioaji每日股價抓取、盤中13:00帶量長紅K掃描
Downstream: —
Transitive downstream (0): —

Edges (typed evidence)

type	hard?	from	to	resource	origin	conf	evidence
DATA	soft	Shioaji每日股價抓取	Prospective Shadow Exit Runner Intraday	sqlite:stock_kline	STATIC	0.75	table 'stock_kline' write@205fea19ed1e(line 53) -> read@b4937395cbb3(line 177) [pure consumer]
DATA	soft	盤中13:00帶量長紅K掃描	Prospective Shadow Exit Runner Intraday	sqlite:stock_kline	STATIC	0.75	table 'stock_kline' write@ce780536ef1c(line 53) -> read@b4937395cbb3(line 177) [pure consumer]

11. Shared resources

stock_kline — shared with 16 jobs
stock_kline — shared with 16 jobs
stock_kline — shared with 16 jobs
+00:00 — shared with 15 jobs
+00:00 — shared with 15 jobs
+00:00 — shared with 15 jobs
sqlite_master — shared with 13 jobs
discovery_intraday_bars — shared with 11 jobs
discovery_intraday_bars — shared with 11 jobs
discovery_intraday_bars — shared with 11 jobs

12. Recent executions

failure_streak=0
last_error=—
delivery_error=—

14. Drift history

2026-09-05T07:09 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on '盤中13:00帶量長紅K掃描' (earliest 13:00) via DATA [sqlite:stock_kline]
2026-09-05T07:09 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on 'Shioaji每日股價抓取' (earliest 15:35) via DATA [sqlite:stock_kline]
2026-09-05T00:38 JOB_ADDED job b4937395cbb3 appeared (Prospective Shadow Exit Runner Intraday)
2026-09-04T23:49 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on '盤中13:00帶量長紅K掃描' (earliest 13:00) via DATA [sqlite:stock_kline]
2026-09-04T23:49 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on 'Shioaji每日股價抓取' (earliest 15:35) via DATA [sqlite:stock_kline]
2026-09-04T23:25 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on '盤中13:00帶量長紅K掃描' (earliest 13:00) via DATA [sqlite:stock_kline]
2026-09-04T23:25 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on 'Shioaji每日股價抓取' (earliest 15:35) via DATA [sqlite:stock_kline]
2026-09-04T22:51 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on '盤中13:00帶量長紅K掃描' (earliest 13:00) via DATA [sqlite:stock_kline]
2026-09-04T22:51 TEMPORAL_DEPENDENCY_HAZARD 'Prospective Shadow Exit Runner Intraday' fires 09:00 but depends on 'Shioaji每日股價抓取' (earliest 15:35) via DATA [sqlite:stock_kline]
2026-09-04T22:51 JOB_ADDED job b4937395cbb3 appeared (Prospective Shadow Exit Runner Intraday)

16. Impact / change risk

change_risk=MEDIUM