Brooks factory state stack

State Layers

Each layer turns a Brooks research finding into a read-only review field. Every card pairs the layer with a real 5-minute session from this week and links the Codex review packet that defined it — so scanner cards, replay notes, dashboards, and journals can explain why a setup is allowed, blocked, downgraded, or still unresolved.

Research only

No live-rule authority

State layers

in current review stack

Newest layers

2 this week

reversal proof · stop pressure

Example data

May 11–15

real 5-minute sessions this week

Publish surface

llms.txt

GitHub AI-readable handoff

Decision flow

A setup label enters late

GitHub knowledge pack

Step 1

Context owns direction

Step 2

Location creates edge

Step 3

Proof confirms or blocks

Step 4

Replay decides promotion

Context

Timeframes

Fractal Timeframe Authority

raw-book-backed

Decides which timeframe owns context and which timeframe is only providing entry detail.

Fields — what each one records

context_tf: The higher timeframe that owns the context.
execution_tf: The timeframe the trade is executed on.
trigger_tf: The timeframe that generates the entry signal.
tf_regime_by_frame: The regime label on each timeframe.
timeframe_authority_state: Which timeframe's read currently wins.

TSM trend-from-open

TSM · May 14 · 5m

Lower-TF dip only

A one-bar dip looks like a reversal up close, but it is only a pullback inside the higher-timeframe bull trend.

Downgrade countertrend expectation unless higher-timeframe damage appears.

Do not average conflicting timeframes into vague confluence.

Codex packet · May 1

Regime

Range

Range Gravity

raw-book-backed

Flags when the market still behaves like a range even after a breakout attempt.

Fields — what each one records

range_gravity_state: How strongly price is still being pulled back into the range.
range_position: Where price sits in the range — top, middle, or bottom.
breakout_conversion_score: Odds a breakout actually sticks.
first_breakout_result: How the first breakout attempt resolved.

MU breakout attempt

MU · May 15 · 5m

Breakout falls back

An early push higher cannot hold — price falls straight back into the prior overlap before the real move.

Treat the first poke as range behavior and wait for acceptance.

Do not trade the middle as if it were an edge.

Codex packet · Apr 29

Channel

Channel Pressure And Pullback Depth

local book-derived

Describes whether a pullback is healthy trend behavior or channel pressure is decaying.

Fields — what each one records

channel_pressure_state: Whether the channel is still pushing or losing force.
pullback_depth_class: How deep the pullback is — shallow, moderate, or deep.
pullback_duration_bars: How many bars the pullback has lasted.
channel_break_status: Whether price has broken the channel line.

COHR bull channel pullback

COHR · May 11 · 5m

Deep pullback

A steady bull channel runs, then a two-bar pullback digs deeper than the earlier shallow tests.

Move from continuation assumption to proof-required review.

A channel line break is not automatically a reversal.

Codex packet · Apr 29

Direction

Always In

Always In Bias State

local book-derived

Records who owns the market before a setup label is allowed to matter.

Fields — what each one records

always_in_state: Which side currently owns the market — long, short, or neutral.
always_in_strength: How strongly that directional bias is held.
always_in_mode: What kind of trend framework is active — spike, channel, or range.
state_alignment: Whether the timeframes agree on direction.

QCOM opening bear drive

QCOM · May 12 · 5m

Sellers reassert

A one-bar bounce off the open fails; the third bar sells through it and every close after stays lower — 5M Always In short.

Review short-continuation entries; treat bounces as exits, not reversals.

Do not turn forced-choice bias into automatic entry permission.

Codex packet · Apr 29

Location

Level Test

Support/Resistance Acceptance Test

raw-book-backed

Treats a level as a test that must be accepted or rejected after contact.

Fields — what each one records

level_id: Identifier for the support/resistance level being tested.
tested_level_type: What kind of level it is — support, resistance, breakout point.
retest_result: Whether the retest held or failed.
trapped_side: Which side got caught on the wrong side of the level.
post_test_follow_through: Price action in the bars after the test.

SPY breakout retest

SPY · May 13 · 5m

Breakout accepted

Price coils against a level for several bars, then a bull bar closes above it and the next bars hold the breakout.

Upgrade continuation review only after the post-test follow-through bar.

A level touch is evidence gathering, not a trigger.

Codex packet · Apr 29

Compression

Breakout Mode

Breakout Mode And Range Middle

raw-book-backed

Marks compression where both sides are waiting for a clear failed breakout or accepted breakout.

Fields — what each one records

compression_type: What kind of tight range price is coiling in.
breakout_mode_state: Whether the market is waiting, breaking out, or failing.
first_breakout_direction: Which way the first breakout attempt went.
failed_failure_result: How a failed breakout's failure resolved.

XLF tight compression

XLF · May 12 · 5m

First breakout fails

Inside a 20-cent range, the first poke above the top fails back before a later breakout finally holds.

Review after the sequence resolves, not on the first poke.

Breakout mode is a wait state until evidence resolves.

Codex packet · Apr 30

Maturity

Leg Count

Leg Count And Second Entry

local book-derived

Tracks whether the market is on a first attempt, second entry, failed second entry, wedge, or TBTL clock.

Fields — what each one records

attempt_number: Which try this is — first attempt, second entry, and so on.
legs_completed_since_spike: How many price legs have run since the last spike.
last_failed_entry_side: Direction of the most recent failed entry.
tbtl_clock_remaining: Bars left before the move is 'too late to trade'.

NVDA first pullback after the open

NVDA · May 13 · 5m

Failed first attempt

The first bounce off the low rolls over and fails; the second attempt is the one that holds and grinds up.

Log the attempt number before grading the setup.

A setup label without attempt number is incomplete.

Codex packet · Apr 29

Evidence

Signal Gate

Signal Context Gate

local book-derived

Separates attractive bar shapes from signals that actually have context and follow-through.

Fields — what each one records

shape_score: How clean the signal bar's shape is on its own.
context_score: How well the signal fits the surrounding context.
entry_bar_score: Quality of the bar you'd actually enter on.
follow_through_score: How convincingly price extended after the entry.
signal_block_reason: Why a signal was rejected, when it was.

MRVL opening spike

MRVL · May 13 · 5m

Entry bar fails

The spike bar tags the session high, but the next bar pokes higher and closes straight back inside it.

Mark as failed proof, not an accepted breakout continuation.

A pretty signal bar is not enough when context or entry-bar proof is missing.

Codex packet · Apr 29

Transition

Climax

Climax Resolution

local book-derived

Separates early spike continuation from late exhaustion and unresolved climax behavior.

Fields — what each one records

climax_resolution_state: How the exhaustion move resolved — continuation or reversal.
climax_phase: Which stage of the climax price is in.
climax_shape: The geometry of the climax — spike, blow-off, or wedge.
exhaustion_context: The conditions leading into the exhaustion.

CRCL late third push

CRCL · May 11 · 5m

Climax push, weak close

The move accelerates into a new high, then the final bar prints the high and closes weak.

Require resolution before buying another late breakout.

Do not turn climax labels into automatic fade rules.

Codex packet · Apr 29

Reversal Proof

Minor Vs Major Reversal Proof State

local book-derived

Prevents first countertrend signals from being promoted into major reversal ideas too early.

Fields — what each one records

reversal_proof_state: How much evidence supports a real reversal.
prior_trend_damage: How badly the prior trend has been broken.
old_extreme_test_type: How the prior high or low was retested.
second_reversal_quality: Strength of the confirming second reversal.

AMZN double bottom

AMZN · May 14 · 5m

Second low

A sharp drop bottoms, then a second low tests the same price — a double bottom with no strong second reversal yet.

Classify as a minor-reversal candidate for replay, not a live signal.

Most first reversals are minor, range-forming, or failed.

Codex packet · May 16new this week

Momentum

Gap

Gap Continuity And Gap Closure

implementation-derived

Tracks whether gaps are supporting continuation or closing into disappointment.

Fields — what each one records

gap_type: What kind of gap this is.
gap_direction: Whether the gap is up or down.
gap_midpoint: The mid price of the gap — a common test level.
gap_continuity_state: Whether gaps are supporting the trend or closing against it.
body_gap_closure_count: How many body gaps have been filled.

F opening continuation

F · May 13 · 5m

Pullback holds the gap

Higher closes stack bar after bar; the one pullback cannot close the gap and momentum resumes.

Treat as momentum evidence that still needs trader-equation review.

A gap or gap close is not automatic entry logic.

Codex packet · Apr 29

Pressure

Stop Pressure

Stop Pressure And Breakeven Test

raw-book-backed

Shows whose protective stop, breakeven price, or failed-entry price is being tested.

Fields — what each one records

stop_pressure_zone: Where protective stops are clustered.
breakeven_test_state: Whether price has come back to test entry cost.
scalper_profit_before_pullback: Whether quick traders were paid before the pullback.
stop_distance_mode: How the stop is sized relative to noise.

MSFT breakout pullback

MSFT · May 15 · 5m

Breakeven test

After the thrust higher, the pullback grinds back to the breakout price and sits on breakout-trader breakevens.

Downgrade the accepted breakout and tag trapped-trader risk.

Stop clusters matter only with regime, location, follow-through, and trader-equation support.

Codex packet · May 16new this week

Management

Order Fill

Order Fill And Missed Target Pressure

raw-book-backed

Checks whether price moved far enough to pay traders who entered, or disappointed them.

Fields — what each one records

order_fill_pressure_state: Whether the move paid entered traders or disappointed them.
target_type: What kind of target was in play — scalp, measured move, etc.
target_fill_status: Whether the target was reached.
missed_target_trap: Price tagged near the target, then reversed.

META push into target

META · May 14 · 5m

Tags target, reverses

Price stretches to within a tick of the prior high, then reverses hard before paying the late entries.

Tag the missed-target disappointment before judging the original entry.

Missed target pressure is context evidence, not a standalone fade.

Codex packet · Apr 29

Review

Decision

Candidate Decision Record

operating-model-only

Keeps gray-zone decisions auditable instead of hiding them as intuition.

Fields — what each one records

setup_name: Which setup pattern this candidate is.
exception_flags: Any rule exceptions that were waved through.
confidence_band: Reliability tier — high, medium, or gray-zone.
trade_class: Risk/reward category of the trade.
skip_reason: Why the trade was passed on, when skipped.

MU mid-range stop entry

MU · May 13 · 5m

Overlap, no edge

Bars overlap heavily around the range middle with no clean edge — a genuine gray-zone candidate.

Record the no-trade reason and revisit in rejected-winner review.

Exception tracking is workflow evidence, not a new source of trading edge.

Codex packet · Apr 29