Prompts to decide and protect focus

Objective: Turn scattered records into a factual map of the week. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: PERIOD=week of August 24–28; RECORDS=five-day calendar with 17 blocks, 12 tasks and one delivered proposal; OUTCOME=proposal reviewed and sent Thursday at 16:00; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Personal Monday-to-Friday record: 14 h of meetings, 11 h of focused work, 6 h of messages, and 4 h of administration. The proposal is due Friday; two blocks have no verifiable output, and three interruptions were marked urgent without a stated consequence.Apply: Time audit; Fact–assumption separation; Verifiable personal review.Order: Gather available calendar, task and note records. → Close the diagnosis with the period you can actually observe.Deliver: Time-use map; Every block has a source or a marker.Boundary: Do not invent periods or outputs.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPERIOD = week of August 24–28RECORDS = five-day calendar with 17 blocks, 12 tasks and one delivered proposalOUTCOME = proposal reviewed and sent Thursday at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Personal Monday-to-Friday record: 14 h of meetings, 11 h of focused work, 6 h of messages, and 4 h of administration. The proposal is due Friday; two blocks have no verifiable output, and three interruptions were marked urgent without a stated consequence.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn scattered records into a factual map of the week. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Time audit- Fact–assumption separation- Verifiable personal review# Workflow1. Gather available calendar, task and note records.2. Order activities by day, duration and output.3. Mark interruptions and missing data.4. Summarize patterns as observations, not causes.5. Close the diagnosis with the period you can actually observe.# Boundaries- Do not invent periods or outputs.- Declare any reconstruction from memory.- Do not fill record gaps with an imagined week.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Describe the observed period; do not predict performance.- It does not delegate personal decisions to AI.# Expected output- Time-use map- Every block has a source or a marker.- The summary separates facts from inference.- At least one relevant gap is identified.- The map identifies which records you can verify yourself.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: When you need a baseline before changing the calendar. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPERIOD = week of August 24–28RECORDS = five-day calendar with 17 blocks, 12 tasks and one delivered proposalOUTCOME = proposal reviewed and sent Thursday at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Personal Monday-to-Friday record: 14 h of meetings, 11 h of focused work, 6 h of messages, and 4 h of administration. The proposal is due Friday; two blocks have no verifiable output, and three interruptions were marked urgent without a stated consequence.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Time-use mapInclude: Turn scattered records into a factual map of the week. Apply the decision to your own practice.Exclude: Describe the observed period; do not predict performance.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Time audit- Fact–assumption separation- Verifiable personal reviewSteps:1. Gather available calendar, task and note records.2. Order activities by day, duration and output.3. Mark interruptions and missing data.4. Summarize patterns as observations, not causes.5. Close the diagnosis with the period you can actually observe.### Edge cases- Incomplete records: deliver a partial map.- Overlapping activities: keep both and mark conflict.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Time-use map- Every block has a source or a marker.- The summary separates facts from inference.- At least one relevant gap is identified.- The map identifies which records you can verify yourself.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every block has a source or a marker.- The summary separates facts from inference.- At least one relevant gap is identified.- The map identifies which records you can verify yourself.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every block has a source or a marker.; The summary separates facts from inference.; At least one relevant gap is identified.; The map identifies which records you can verify yourself.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: More detail improves traceability but increases recording effort. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-01- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Time audit- Fact–assumption separation- Verifiable personal review# Boundaries- Do not invent periods or outputs.- Declare any reconstruction from memory.- Do not fill record gaps with an imagined week.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Describe the observed period; do not predict performance.- It does not delegate personal decisions to AI.# Definition of DoneEvery block has a source or a marker.; The summary separates facts from inference.; At least one relevant gap is identified.; The map identifies which records you can verify yourself.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPERIOD = week of August 24–28RECORDS = five-day calendar with 17 blocks, 12 tasks and one delivered proposalOUTCOME = proposal reviewed and sent Thursday at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Personal Monday-to-Friday record: 14 h of meetings, 11 h of focused work, 6 h of messages, and 4 h of administration. The proposal is due Friday; two blocks have no verifiable output, and three interruptions were marked urgent without a stated consequence.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn scattered records into a factual map of the week. Apply the decision to your own practice.# Workflow1. Gather available calendar, task and note records.2. Order activities by day, duration and output.3. Mark interruptions and missing data.4. Summarize patterns as observations, not causes.5. Close the diagnosis with the period you can actually observe.# Expected output- Time-use map- Every block has a source or a marker.- The summary separates facts from inference.- At least one relevant gap is identified.- The map identifies which records you can verify yourself.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Classify activities by their relationship to value. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: ACTIVITIES=6 meetings, 4 analysis hours and 3 writing hours; EFFORT=meetings 6 h; analysis 4 h; writing 3 h; VALUE_SIGNAL=the review unlocked a decision; two meetings produced no output; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Factual map: 35 h recorded; proposal 6 h and one draft; meetings 14 h and three decisions; messages 6 h with no deliverable; administration 4 h; 5 h unrecorded. Expected weekly outcome: proposal reviewed on Thursday.Apply: Value map; Result job-to-be-done; Verifiable personal review.Order: Read the period purpose and results. → Personally choose which activity to keep, reduce or review.Deliver: Activity value map; Every activity has a category and reason.Boundary: Do not trivialize necessary maintenance.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSACTIVITIES = 6 meetings, 4 analysis hours and 3 writing hoursEFFORT = meetings 6 h; analysis 4 h; writing 3 hVALUE_SIGNAL = the review unlocked a decision; two meetings produced no outputPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Factual map: 35 h recorded; proposal 6 h and one draft; meetings 14 h and three decisions; messages 6 h with no deliverable; administration 4 h; 5 h unrecorded. Expected weekly outcome: proposal reviewed on Thursday.# Optional adjustmentsState the missing fact that would change the decision.# TaskClassify activities by their relationship to value. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Value map- Result job-to-be-done- Verifiable personal review# Workflow1. Read the period purpose and results.2. Classify creation, maintenance, coordination or reaction.3. Justify ambiguous cases.4. Deliver decisions and open questions.5. Personally choose which activity to keep, reduce or review.# Boundaries- Do not trivialize necessary maintenance.- Do not assign value without a stated purpose.- Do not classify necessary personal duties as waste.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Classification is contextual, not universal.- It does not delegate personal decisions to AI.# Expected output- Activity value map- Every activity has a category and reason.- Uncertain cases preserve missing evidence.- The map supports keep, reduce or review decisions.- The final decision states which classification you will personally review.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: When the time map exists and you must decide what to question. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSACTIVITIES = 6 meetings, 4 analysis hours and 3 writing hoursEFFORT = meetings 6 h; analysis 4 h; writing 3 hVALUE_SIGNAL = the review unlocked a decision; two meetings produced no outputPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Factual map: 35 h recorded; proposal 6 h and one draft; meetings 14 h and three decisions; messages 6 h with no deliverable; administration 4 h; 5 h unrecorded. Expected weekly outcome: proposal reviewed on Thursday.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Activity value mapInclude: Classify activities by their relationship to value. Apply the decision to your own practice.Exclude: Classification is contextual, not universal.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Value map- Result job-to-be-done- Verifiable personal reviewSteps:1. Read the period purpose and results.2. Classify creation, maintenance, coordination or reaction.3. Justify ambiguous cases.4. Deliver decisions and open questions.5. Personally choose which activity to keep, reduce or review.### Edge cases- One activity serves two categories: record primary and secondary.- If a constraint cannot move, adapt the plan instead of hiding it.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Activity value map- Every activity has a category and reason.- Uncertain cases preserve missing evidence.- The map supports keep, reduce or review decisions.- The final decision states which classification you will personally review.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every activity has a category and reason.- Uncertain cases preserve missing evidence.- The map supports keep, reduce or review decisions.- The final decision states which classification you will personally review.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every activity has a category and reason.; Uncertain cases preserve missing evidence.; The map supports keep, reduce or review decisions.; The final decision states which classification you will personally review.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: A simple taxonomy aids decisions but loses nuance; preserve exceptions. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-02- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Value map- Result job-to-be-done- Verifiable personal review# Boundaries- Do not trivialize necessary maintenance.- Do not assign value without a stated purpose.- Do not classify necessary personal duties as waste.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Classification is contextual, not universal.- It does not delegate personal decisions to AI.# Definition of DoneEvery activity has a category and reason.; Uncertain cases preserve missing evidence.; The map supports keep, reduce or review decisions.; The final decision states which classification you will personally review.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSACTIVITIES = 6 meetings, 4 analysis hours and 3 writing hoursEFFORT = meetings 6 h; analysis 4 h; writing 3 hVALUE_SIGNAL = the review unlocked a decision; two meetings produced no outputPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Factual map: 35 h recorded; proposal 6 h and one draft; meetings 14 h and three decisions; messages 6 h with no deliverable; administration 4 h; 5 h unrecorded. Expected weekly outcome: proposal reviewed on Thursday.# Optional adjustmentsState the missing fact that would change the decision.# TaskClassify activities by their relationship to value. Apply the decision to your own practice.# Workflow1. Read the period purpose and results.2. Classify creation, maintenance, coordination or reaction.3. Justify ambiguous cases.4. Deliver decisions and open questions.5. Personally choose which activity to keep, reduce or review.# Expected output- Activity value map- Every activity has a category and reason.- Uncertain cases preserve missing evidence.- The map supports keep, reduce or review decisions.- The final decision states which classification you will personally review.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Choose three results and expose trade-offs. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: CANDIDATES=close proposal, answer clients and update report; CRITERIA=delivery impact, due date and remaining effort; CAPACITY=12 protectable hours and two required meetings; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Activity classification: proposal = creation; scope meeting = coordination; daily status = maintenance; reactive messages = reaction. Two meetings produced no decision. Candidate outcomes: reviewed proposal, budget decision, and closure of one critical source.Apply: Priority matrix; Definition of done; Verifiable personal review.Order: Write the definition of done. → State the weekly trade-off you can sustain.Deliver: Three justified priorities; Three observable results.Boundary: Do not use pressure as the only criterion.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSCANDIDATES = close proposal, answer clients and update reportCRITERIA = delivery impact, due date and remaining effortCAPACITY = 12 protectable hours and two required meetingsPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Activity classification: proposal = creation; scope meeting = coordination; daily status = maintenance; reactive messages = reaction. Two meetings produced no decision. Candidate outcomes: reviewed proposal, budget decision, and closure of one critical source.# Optional adjustmentsState the missing fact that would change the decision.# TaskChoose three results and expose trade-offs. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Priority matrix- Definition of done- Verifiable personal review# Workflow1. Write the definition of done.2. Compare purpose, impact and real urgency.3. Declare deferred work.4. Deliver three justified priorities.5. State the weekly trade-off you can sustain.# Boundaries- Do not use pressure as the only criterion.- Expose data that could change the order.- Do not accept a priority that conflicts with a non-negotiable boundary.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It does not guarantee completion; it creates a reviewable decision.- It does not delegate personal decisions to AI.# Expected output- Three justified priorities- Three observable results.- Each has a rationale and trade-off.- No hidden tie remains.- Every priority fits within a stated personal boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: When candidate results exceed available capacity. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSCANDIDATES = close proposal, answer clients and update reportCRITERIA = delivery impact, due date and remaining effortCAPACITY = 12 protectable hours and two required meetingsPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Activity classification: proposal = creation; scope meeting = coordination; daily status = maintenance; reactive messages = reaction. Two meetings produced no decision. Candidate outcomes: reviewed proposal, budget decision, and closure of one critical source.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Three justified prioritiesInclude: Choose three results and expose trade-offs. Apply the decision to your own practice.Exclude: It does not guarantee completion; it creates a reviewable decision.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Priority matrix- Definition of done- Verifiable personal reviewSteps:1. Write the definition of done.2. Compare purpose, impact and real urgency.3. Declare deferred work.4. Deliver three justified priorities.5. State the weekly trade-off you can sustain.### Edge cases- Everything seems urgent: request evidence of date and consequence.- Capacity is missing: reduce scope before adding priority.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Three justified priorities- Three observable results.- Each has a rationale and trade-off.- No hidden tie remains.- Every priority fits within a stated personal boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Three observable results.- Each has a rationale and trade-off.- No hidden tie remains.- Every priority fits within a stated personal boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Three observable results.; Each has a rationale and trade-off.; No hidden tie remains.; Every priority fits within a stated personal boundary.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: A limit of three creates clarity but defers legitimate work. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-03- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Priority matrix- Definition of done- Verifiable personal review# Boundaries- Do not use pressure as the only criterion.- Expose data that could change the order.- Do not accept a priority that conflicts with a non-negotiable boundary.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It does not guarantee completion; it creates a reviewable decision.- It does not delegate personal decisions to AI.# Definition of DoneThree observable results.; Each has a rationale and trade-off.; No hidden tie remains.; Every priority fits within a stated personal boundary.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSCANDIDATES = close proposal, answer clients and update reportCRITERIA = delivery impact, due date and remaining effortCAPACITY = 12 protectable hours and two required meetingsPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Activity classification: proposal = creation; scope meeting = coordination; daily status = maintenance; reactive messages = reaction. Two meetings produced no decision. Candidate outcomes: reviewed proposal, budget decision, and closure of one critical source.# Optional adjustmentsState the missing fact that would change the decision.# TaskChoose three results and expose trade-offs. Apply the decision to your own practice.# Workflow1. Write the definition of done.2. Compare purpose, impact and real urgency.3. Declare deferred work.4. Deliver three justified priorities.5. State the weekly trade-off you can sustain.# Expected output- Three justified priorities- Three observable results.- Each has a rationale and trade-off.- No hidden tie remains.- Every priority fits within a stated personal boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Test priorities against dependencies, effort, risk and calendar reality. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: PRIORITIES=1 proposal; 2 client reply; 3 report; DEPENDENCIES=financial data Tuesday and legal review Wednesday; RISKS=scope change or delayed legal review; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Three prioritized outcomes: proposal approved Thursday; budget decision Friday; source rights confirmed Wednesday. Real capacity: 10 h. Reviewer available Thursday at 15:00. Supplier dependency still unconfirmed. Boundary: no work after 18:00.Apply: Short pre-mortem; Impact–effort–risk matrix; Verifiable personal review.Order: List assumptions and dependencies. → Decide which risk you will accept and which needs external help.Deliver: Visible risks and dependencies; Every risk has a signal and response.Boundary: Do not invent availability or agreements.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportDEPENDENCIES = financial data Tuesday and legal review WednesdayRISKS = scope change or delayed legal reviewPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Three prioritized outcomes: proposal approved Thursday; budget decision Friday; source rights confirmed Wednesday. Real capacity: 10 h. Reviewer available Thursday at 15:00. Supplier dependency still unconfirmed. Boundary: no work after 18:00.# Optional adjustmentsState the missing fact that would change the decision.# TaskTest priorities against dependencies, effort, risk and calendar reality. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Short pre-mortem- Impact–effort–risk matrix- Verifiable personal review# Workflow1. List assumptions and dependencies.2. Compare effort with declared available space.3. Identify risk and mitigation.4. Recommend keep, adjust or remove.5. Decide which risk you will accept and which needs external help.# Boundaries- Do not invent availability or agreements.- Separate internal from external dependency.- Do not assume personal control over an external dependency.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- This is a reality test, not certainty about the future.- It does not delegate personal decisions to AI.# Expected output- Visible risks and dependencies- Every risk has a signal and response.- Dependencies have status.- The final decision explains the trade-off.- The recommendation separates your action from required help.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: Before reserving blocks or committing a date. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportDEPENDENCIES = financial data Tuesday and legal review WednesdayRISKS = scope change or delayed legal reviewPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Three prioritized outcomes: proposal approved Thursday; budget decision Friday; source rights confirmed Wednesday. Real capacity: 10 h. Reviewer available Thursday at 15:00. Supplier dependency still unconfirmed. Boundary: no work after 18:00.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Visible risks and dependenciesInclude: Test priorities against dependencies, effort, risk and calendar reality. Apply the decision to your own practice.Exclude: This is a reality test, not certainty about the future.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Short pre-mortem- Impact–effort–risk matrix- Verifiable personal reviewSteps:1. List assumptions and dependencies.2. Compare effort with declared available space.3. Identify risk and mitigation.4. Recommend keep, adjust or remove.5. Decide which risk you will accept and which needs external help.### Edge cases- Dependency has no owner: mark blocked.- Risk lacks evidence: treat as hypothesis.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Visible risks and dependencies- Every risk has a signal and response.- Dependencies have status.- The final decision explains the trade-off.- The recommendation separates your action from required help.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every risk has a signal and response.- Dependencies have status.- The final decision explains the trade-off.- The recommendation separates your action from required help.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every risk has a signal and response.; Dependencies have status.; The final decision explains the trade-off.; The recommendation separates your action from required help.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: More validation reduces surprise but consumes time before execution. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-04- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Short pre-mortem- Impact–effort–risk matrix- Verifiable personal review# Boundaries- Do not invent availability or agreements.- Separate internal from external dependency.- Do not assume personal control over an external dependency.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- This is a reality test, not certainty about the future.- It does not delegate personal decisions to AI.# Definition of DoneEvery risk has a signal and response.; Dependencies have status.; The final decision explains the trade-off.; The recommendation separates your action from required help.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportDEPENDENCIES = financial data Tuesday and legal review WednesdayRISKS = scope change or delayed legal reviewPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Three prioritized outcomes: proposal approved Thursday; budget decision Friday; source rights confirmed Wednesday. Real capacity: 10 h. Reviewer available Thursday at 15:00. Supplier dependency still unconfirmed. Boundary: no work after 18:00.# Optional adjustmentsState the missing fact that would change the decision.# TaskTest priorities against dependencies, effort, risk and calendar reality. Apply the decision to your own practice.# Workflow1. List assumptions and dependencies.2. Compare effort with declared available space.3. Identify risk and mitigation.4. Recommend keep, adjust or remove.5. Decide which risk you will accept and which needs external help.# Expected output- Visible risks and dependencies- Every risk has a signal and response.- Dependencies have status.- The final decision explains the trade-off.- The recommendation separates your action from required help.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Turn priorities into blocks with output and boundary. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: PRIORITIES=1 proposal; 2 client reply; 3 report; CALENDAR=Tuesday and Wednesday 09:00–11:00 free; meetings in the afternoon; CONSTRAINTS=required meetings, Thursday delivery and at most two overtime hours; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Reality test: keep proposal scope at version 1; reduce the budget memo to one page; defer dashboard polish. Estimate: 3 h for inputs and 2 h for review. Risk: reviewer availability. Mitigation: send the draft Wednesday at noon.Apply: Time blocking; Definition of done; Verifiable personal review.Order: Assign blocks using declared work and available energy. → Close with a calendar compatible with your real commitments.Deliver: Protected weekly calendar; Every priority has a block.Boundary: Do not fill all capacity.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportCALENDAR = Tuesday and Wednesday 09:00–11:00 free; meetings in the afternoonCONSTRAINTS = required meetings, Thursday delivery and at most two overtime hoursPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Reality test: keep proposal scope at version 1; reduce the budget memo to one page; defer dashboard polish. Estimate: 3 h for inputs and 2 h for review. Risk: reviewer availability. Mitigation: send the draft Wednesday at noon.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn priorities into blocks with output and boundary. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Time blocking- Definition of done- Verifiable personal review# Workflow1. Assign blocks using declared work and available energy.2. Define input, output and definition of done.3. Keep coordination margin.4. Deliver calendar and conflicts.5. Close with a calendar compatible with your real commitments.# Boundaries- Do not fill all capacity.- Do not move fixed commitments without authorization.- Do not reserve capacity already committed elsewhere.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- The calendar is an executable hypothesis and must be reviewed.- It does not delegate personal decisions to AI.# Expected output- Protected weekly calendar- Every priority has a block.- Every block has an output.- Conflicts are marked.- The calendar’s next action is under your control.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: When priorities have passed the reality test. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportCALENDAR = Tuesday and Wednesday 09:00–11:00 free; meetings in the afternoonCONSTRAINTS = required meetings, Thursday delivery and at most two overtime hoursPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Reality test: keep proposal scope at version 1; reduce the budget memo to one page; defer dashboard polish. Estimate: 3 h for inputs and 2 h for review. Risk: reviewer availability. Mitigation: send the draft Wednesday at noon.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Protected weekly calendarInclude: Turn priorities into blocks with output and boundary. Apply the decision to your own practice.Exclude: The calendar is an executable hypothesis and must be reviewed.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Time blocking- Definition of done- Verifiable personal reviewSteps:1. Assign blocks using declared work and available energy.2. Define input, output and definition of done.3. Keep coordination margin.4. Deliver calendar and conflicts.5. Close with a calendar compatible with your real commitments.### Edge cases- No block is large enough: split the output.- Dependency pending: schedule preparation, not execution.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Protected weekly calendar- Every priority has a block.- Every block has an output.- Conflicts are marked.- The calendar’s next action is under your control.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every priority has a block.- Every block has an output.- Conflicts are marked.- The calendar’s next action is under your control.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every priority has a block.; Every block has an output.; Conflicts are marked.; The calendar’s next action is under your control.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: Long blocks reduce switching but reduce flexibility for exceptions. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-05- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Time blocking- Definition of done- Verifiable personal review# Boundaries- Do not fill all capacity.- Do not move fixed commitments without authorization.- Do not reserve capacity already committed elsewhere.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- The calendar is an executable hypothesis and must be reviewed.- It does not delegate personal decisions to AI.# Definition of DoneEvery priority has a block.; Every block has an output.; Conflicts are marked.; The calendar’s next action is under your control.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSPRIORITIES = 1 proposal; 2 client reply; 3 reportCALENDAR = Tuesday and Wednesday 09:00–11:00 free; meetings in the afternoonCONSTRAINTS = required meetings, Thursday delivery and at most two overtime hoursPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Reality test: keep proposal scope at version 1; reduce the budget memo to one page; defer dashboard polish. Estimate: 3 h for inputs and 2 h for review. Risk: reviewer availability. Mitigation: send the draft Wednesday at noon.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn priorities into blocks with output and boundary. Apply the decision to your own practice.# Workflow1. Assign blocks using declared work and available energy.2. Define input, output and definition of done.3. Keep coordination margin.4. Deliver calendar and conflicts.5. Close with a calendar compatible with your real commitments.# Expected output- Protected weekly calendar- Every priority has a block.- Every block has an output.- Conflicts are marked.- The calendar’s next action is under your control.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Define default responses for interruptions. Apply the decision to your own practice. Apply the protocol for one practice week and keep the record that prompt-07 will review. The final decision and evidence remain under your responsibility.Data: INTERRUPTIONS=11 urgent messages and two ad hoc meetings in five days; RULES=check messages at 12:00 and 16:30; escalate only critical blockers; EXCEPTIONS=client incident with immediate delivery impact; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Proposed blocks: Tuesday 09:00–11:00 outline; Wednesday 09:00–12:00 draft; Thursday 10:00–11:00 verification; Thursday 15:00–16:00 review; Friday 09:00–10:00 decision. Two hours remain unassigned as reserve. Conflict: meeting Wednesday at 10:00.Apply: Interruption protocol; Escalation rules; Verifiable personal review.Order: Group interruption types. → Select one personal response for each recurring interruption.Deliver: Focus protection rules; Every type has response and timing.Boundary: Include a path for real urgency.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSINTERRUPTIONS = 11 urgent messages and two ad hoc meetings in five daysRULES = check messages at 12:00 and 16:30; escalate only critical blockersEXCEPTIONS = client incident with immediate delivery impactPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Proposed blocks: Tuesday 09:00–11:00 outline; Wednesday 09:00–12:00 draft; Thursday 10:00–11:00 verification; Thursday 15:00–16:00 review; Friday 09:00–10:00 decision. Two hours remain unassigned as reserve. Conflict: meeting Wednesday at 10:00.# Optional adjustmentsState the missing fact that would change the decision.# TaskDefine default responses for interruptions. Apply the decision to your own practice. Apply the protocol for one practice week and keep the record that prompt-07 will review.# FRAMEWORKS AND BEST PRACTICES- Interruption protocol- Escalation rules- Verifiable personal review# Workflow1. Group interruption types.2. Assign handle, delegate, batch, defer or discard.3. Define exception and channel.4. Draft concise responses.5. Select one personal response for each recurring interruption.# Boundaries- Include a path for real urgency.- Do not block care, safety or critical service duties.- Do not use focus as an excuse to ignore legitimate urgency.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Adapt the protocol to context and review its effects.- It does not delegate personal decisions to AI.# Expected output- Focus protection rules- Every type has response and timing.- Exceptions are explicit.- Language protects work without hostility.- Every proposed response respects your real responsibilities.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: When messages and requests repeatedly displace blocks. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSINTERRUPTIONS = 11 urgent messages and two ad hoc meetings in five daysRULES = check messages at 12:00 and 16:30; escalate only critical blockersEXCEPTIONS = client incident with immediate delivery impactPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Proposed blocks: Tuesday 09:00–11:00 outline; Wednesday 09:00–12:00 draft; Thursday 10:00–11:00 verification; Thursday 15:00–16:00 review; Friday 09:00–10:00 decision. Two hours remain unassigned as reserve. Conflict: meeting Wednesday at 10:00.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Focus protection rulesInclude: Define default responses for interruptions. Apply the decision to your own practice. Apply the protocol for one practice week and keep the record that prompt-07 will review.Exclude: Adapt the protocol to context and review its effects.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Interruption protocol- Escalation rules- Verifiable personal reviewSteps:1. Group interruption types.2. Assign handle, delegate, batch, defer or discard.3. Define exception and channel.4. Draft concise responses.5. Select one personal response for each recurring interruption.### Edge cases- Ambiguous request: ask for consequence and date.- Authority conflict: escalate instead of improvising.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Focus protection rules- Every type has response and timing.- Exceptions are explicit.- Language protects work without hostility.- Every proposed response respects your real responsibilities.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every type has response and timing.- Exceptions are explicit.- Language protects work without hostility.- Every proposed response respects your real responsibilities.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every type has response and timing.; Exceptions are explicit.; Language protects work without hostility.; Every proposed response respects your real responsibilities.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: A rule protects focus but can delay coordination without exceptions. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-06- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Interruption protocol- Escalation rules- Verifiable personal review# Boundaries- Include a path for real urgency.- Do not block care, safety or critical service duties.- Do not use focus as an excuse to ignore legitimate urgency.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Adapt the protocol to context and review its effects.- It does not delegate personal decisions to AI.# Definition of DoneEvery type has response and timing.; Exceptions are explicit.; Language protects work without hostility.; Every proposed response respects your real responsibilities.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSINTERRUPTIONS = 11 urgent messages and two ad hoc meetings in five daysRULES = check messages at 12:00 and 16:30; escalate only critical blockersEXCEPTIONS = client incident with immediate delivery impactPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Proposed blocks: Tuesday 09:00–11:00 outline; Wednesday 09:00–12:00 draft; Thursday 10:00–11:00 verification; Thursday 15:00–16:00 review; Friday 09:00–10:00 decision. Two hours remain unassigned as reserve. Conflict: meeting Wednesday at 10:00.# Optional adjustmentsState the missing fact that would change the decision.# TaskDefine default responses for interruptions. Apply the decision to your own practice. Apply the protocol for one practice week and keep the record that prompt-07 will review.# Workflow1. Group interruption types.2. Assign handle, delegate, batch, defer or discard.3. Define exception and channel.4. Draft concise responses.5. Select one personal response for each recurring interruption.# Expected output- Focus protection rules- Every type has response and timing.- Exceptions are explicit.- Language protects work without hostility.- Every proposed response respects your real responsibilities.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Compare plan, execution and evidence to choose one adjustment. Apply the decision to your own practice. Use prompt-08 only when an open decision requires governed sources; otherwise close the review. The final decision and evidence remain under your responsibility.Data: OUTCOME=proposal reviewed and sent Thursday at 16:00; EVIDENCE=five-day board, three citations and artifact version 0.2; OPEN_DECISIONS=retain focus blocks and adjust the response agreement; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Protocol applied for one week: batch messages at 12:00 and 16:00; urgent only for safety or a deadline under 4 h. Log: 11 interruptions, 6 queued, 3 handled, and 2 escalated. The proposal was delivered Thursday; one response missed its agreed deadline.Apply: After Action Review; Plan–fact–learning–decision; Verifiable personal review.Order: Compare results and blocks. → Choose one personal rule for the next cycle.Deliver: Fact-based weekly review; Facts and inferences are separate.Boundary: Do not turn perceived effort into evidence.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSOUTCOME = proposal reviewed and sent Thursday at 16:00EVIDENCE = five-day board, three citations and artifact version 0.2OPEN_DECISIONS = retain focus blocks and adjust the response agreementPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Protocol applied for one week: batch messages at 12:00 and 16:00; urgent only for safety or a deadline under 4 h. Log: 11 interruptions, 6 queued, 3 handled, and 2 escalated. The proposal was delivered Thursday; one response missed its agreed deadline.# Optional adjustmentsState the missing fact that would change the decision.# TaskCompare plan, execution and evidence to choose one adjustment. Apply the decision to your own practice. Use prompt-08 only when an open decision requires governed sources; otherwise close the review.# FRAMEWORKS AND BEST PRACTICES- After Action Review- Plan–fact–learning–decision- Verifiable personal review# Workflow1. Compare results and blocks.2. Describe deviations with facts.3. Identify one useful rule and one friction.4. Propose an adjustment and review date.5. Choose one personal rule for the next cycle.# Boundaries- Do not turn perceived effort into evidence.- Do not generalize from one period.- Do not turn perceived effort into evidence of progress.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Review supports decisions; it does not prove general improvement.- It does not delegate personal decisions to AI.# Expected output- Fact-based weekly review- Facts and inferences are separate.- Each adjustment addresses a friction.- A keep-or-remove criterion exists.- The chosen adjustment has evidence you can record again.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: At period close or before planning the next week. Use records you can review. The sequence may end here.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSOUTCOME = proposal reviewed and sent Thursday at 16:00EVIDENCE = five-day board, three citations and artifact version 0.2OPEN_DECISIONS = retain focus blocks and adjust the response agreementPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Protocol applied for one week: batch messages at 12:00 and 16:00; urgent only for safety or a deadline under 4 h. Log: 11 interruptions, 6 queued, 3 handled, and 2 escalated. The proposal was delivered Thursday; one response missed its agreed deadline.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Fact-based weekly reviewInclude: Compare plan, execution and evidence to choose one adjustment. Apply the decision to your own practice. Use prompt-08 only when an open decision requires governed sources; otherwise close the review.Exclude: Review supports decisions; it does not prove general improvement.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- After Action Review- Plan–fact–learning–decision- Verifiable personal reviewSteps:1. Compare results and blocks.2. Describe deviations with facts.3. Identify one useful rule and one friction.4. Propose an adjustment and review date.5. Choose one personal rule for the next cycle.### Edge cases- Insufficient evidence: preserve the question; do not invent a conclusion.- If a constraint cannot move, adapt the plan instead of hiding it.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Fact-based weekly review- Facts and inferences are separate.- Each adjustment addresses a friction.- A keep-or-remove criterion exists.- The chosen adjustment has evidence you can record again.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Facts and inferences are separate.- Each adjustment addresses a friction.- A keep-or-remove criterion exists.- The chosen adjustment has evidence you can record again.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Facts and inferences are separate.; Each adjustment addresses a friction.; A keep-or-remove criterion exists.; The chosen adjustment has evidence you can record again.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: Testing one rule slows the pace of change but makes the evidence easier to connect to that adjustment. Autonomy preserves context but requires honest self-review. Consulting sources adds traceability but is not required when local evidence is enough to close the decision.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-07- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- After Action Review- Plan–fact–learning–decision- Verifiable personal review# Boundaries- Do not turn perceived effort into evidence.- Do not generalize from one period.- Do not turn perceived effort into evidence of progress.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Review supports decisions; it does not prove general improvement.- It does not delegate personal decisions to AI.# Definition of DoneFacts and inferences are separate.; Each adjustment addresses a friction.; A keep-or-remove criterion exists.; The chosen adjustment has evidence you can record again.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSOUTCOME = proposal reviewed and sent Thursday at 16:00EVIDENCE = five-day board, three citations and artifact version 0.2OPEN_DECISIONS = retain focus blocks and adjust the response agreementPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Protocol applied for one week: batch messages at 12:00 and 16:00; urgent only for safety or a deadline under 4 h. Log: 11 interruptions, 6 queued, 3 handled, and 2 escalated. The proposal was delivered Thursday; one response missed its agreed deadline.# Optional adjustmentsState the missing fact that would change the decision.# TaskCompare plan, execution and evidence to choose one adjustment. Apply the decision to your own practice. Use prompt-08 only when an open decision requires governed sources; otherwise close the review.# Workflow1. Compare results and blocks.2. Describe deviations with facts.3. Identify one useful rule and one friction.4. Propose an adjustment and review date.5. Choose one personal rule for the next cycle.# Expected output- Fact-based weekly review- Facts and inferences are separate.- Each adjustment addresses a friction.- A keep-or-remove criterion exists.- The chosen adjustment has evidence you can record again.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Build a cited synthesis for one concrete decision. Apply the decision to your own practice. The final decision and evidence remain under your responsibility.Data: DECISION=repeat protected blocks for seven days; SOURCES=dated official guide and synthetic five-day record; ACCEPTANCE_CRITERION=each recommendation cites a source, states fit and proposes a test; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Weekly review: proposal completed, budget decision partial, and source rights still unconfirmed. Chosen adjustment: protect Wednesday's drafting block. Open decision: evaluate the practice by time released or quality. Available sources: brief v2, weekly log, and official guide; none establishes causality.Apply: Source triangulation; Claim–evidence–limit table; Verifiable personal review.Order: Define decision and criteria. → Deliver a conditional recommendation.Deliver: Cited decision brief; Every verifiable claim has a citation.Boundary: Stop when relevant sources are absent.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSDECISION = repeat protected blocks for seven daysSOURCES = dated official guide and synthetic five-day recordACCEPTANCE_CRITERION = each recommendation cites a source, states fit and proposes a testPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Weekly review: proposal completed, budget decision partial, and source rights still unconfirmed. Chosen adjustment: protect Wednesday's drafting block. Open decision: evaluate the practice by time released or quality. Available sources: brief v2, weekly log, and official guide; none establishes causality.# Optional adjustmentsState the missing fact that would change the decision.# TaskBuild a cited synthesis for one concrete decision. Apply the decision to your own practice.# FRAMEWORKS AND BEST PRACTICES- Source triangulation- Claim–evidence–limit table- Verifiable personal review# Workflow1. Define decision and criteria.2. Query only available sources.3. Extract claim, citation and limit.4. Compare agreement, tension and gaps.5. Deliver a conditional recommendation.# Boundaries- Stop when relevant sources are absent.- Do not present editorial synthesis as observed fact.- Do not use a source to decide beyond its stated scope.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Use only governed sources available in the notebook.- It does not delegate personal decisions to AI.# Expected output- Cited decision brief- Every verifiable claim has a citation.- Disagreements remain visible.- The recommendation states its limit.- The synthesis exposes the decision that still belongs to you.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: In NotebookLM Sources, when a priority depends on governed information. Use records you can review.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSDECISION = repeat protected blocks for seven daysSOURCES = dated official guide and synthetic five-day recordACCEPTANCE_CRITERION = each recommendation cites a source, states fit and proposes a testPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Weekly review: proposal completed, budget decision partial, and source rights still unconfirmed. Chosen adjustment: protect Wednesday's drafting block. Open decision: evaluate the practice by time released or quality. Available sources: brief v2, weekly log, and official guide; none establishes causality.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Cited decision briefInclude: Build a cited synthesis for one concrete decision. Apply the decision to your own practice.Exclude: Use only governed sources available in the notebook.; It does not delegate personal decisions to AI.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Source triangulation- Claim–evidence–limit table- Verifiable personal reviewSteps:1. Define decision and criteria.2. Query only available sources.3. Extract claim, citation and limit.4. Compare agreement, tension and gaps.5. Deliver a conditional recommendation.### Edge cases- Sources conflict: do not force consensus.- Unclear rights: exclude the source and record the gap.- Incomplete week: return a partial reading and mark the missing period.## C — Verifiable criteria# Expected output:- Cited decision brief- Every verifiable claim has a citation.- Disagreements remain visible.- The recommendation states its limit.- The synthesis exposes the decision that still belongs to you.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every verifiable claim has a citation.- Disagreements remain visible.- The recommendation states its limit.- The synthesis exposes the decision that still belongs to you.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every verifiable claim has a citation.; Disagreements remain visible.; The recommendation states its limit.; The synthesis exposes the decision that still belongs to you.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: Citations improve traceability but may leave an immediate decision unanswered. Autonomy preserves context but requires honest self-review.## Provenance- Synthetic demo- Declared authorities: ev-m02-official-pdf, ev-m02-masterclass, ev-m02-playbook## Metadata- module: 02- prompt: prompt-08- surface: source_searchReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Source triangulation- Claim–evidence–limit table- Verifiable personal review# Boundaries- Stop when relevant sources are absent.- Do not present editorial synthesis as observed fact.- Do not use a source to decide beyond its stated scope.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- Use only governed sources available in the notebook.- It does not delegate personal decisions to AI.# Definition of DoneEvery verifiable claim has a citation.; Disagreements remain visible.; The recommendation states its limit.; The synthesis exposes the decision that still belongs to you.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = executiveAPPROVAL = human required# INPUTSDECISION = repeat protected blocks for seven daysSOURCES = dated official guide and synthetic five-day recordACCEPTANCE_CRITERION = each recommendation cites a source, states fit and proposes a testPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Weekly review: proposal completed, budget decision partial, and source rights still unconfirmed. Chosen adjustment: protect Wednesday's drafting block. Open decision: evaluate the practice by time released or quality. Available sources: brief v2, weekly log, and official guide; none establishes causality.# Optional adjustmentsState the missing fact that would change the decision.# TaskBuild a cited synthesis for one concrete decision. Apply the decision to your own practice.# Workflow1. Define decision and criteria.2. Query only available sources.3. Extract claim, citation and limit.4. Compare agreement, tension and gaps.5. Deliver a conditional recommendation.# Expected output- Cited decision brief- Every verifiable claim has a citation.- Disagreements remain visible.- The recommendation states its limit.- The synthesis exposes the decision that still belongs to you.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Compare the observed weekly practice with relevant sources without replacing local records with generic advice. The final decision and evidence remain under your responsibility.Data: WEEKLY_REVIEW=proposal closed; messages batched; one reply late; SOURCES=dated official guide and synthetic five-day record; APPLICABILITY=individual work with two response windows and a weekly delivery; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Synthetic weekly review: the proposal finished on Thursday; batching messages reduced observed interruptions, but one reply missed its deadline. The decision is to cross-check protected blocks and response agreements before repeating.Apply: Source hierarchy; Evidence-to-decision; Adapt before adopting.Order: Frame a testable question from the weekly review. → Propose one bounded test for the next week.Deliver: Practice cross-check note; Every recommendation has a source and explicit applicability.Boundary: Do not turn general guidance into a proven cause.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSWEEKLY_REVIEW = proposal closed; messages batched; one reply lateSOURCES = dated official guide and synthetic five-day recordAPPLICABILITY = individual work with two response windows and a weekly deliveryPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic weekly review: the proposal finished on Thursday; batching messages reduced observed interruptions, but one reply missed its deadline. The decision is to cross-check protected blocks and response agreements before repeating.# Optional adjustmentsState the missing fact that would change the decision.# TaskCompare the observed weekly practice with relevant sources without replacing local records with generic advice.# FRAMEWORKS AND BEST PRACTICES- Source hierarchy- Evidence-to-decision- Adapt before adopting# Workflow1. Frame a testable question from the weekly review.2. Prioritize primary or dated authoritative sources.3. Compare external guidance with the observed record.4. State fit, conflict and missing evidence.5. Propose one bounded test for the next week.# Boundaries- Do not turn general guidance into a proven cause.- Do not hide differences between the source context and the observed week.- State rights, date and boundary for every source.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It does not prove causality or guarantee improvement.# Expected output- Practice cross-check note- Every recommendation has a source and explicit applicability.- The comparison separates agreement, tension and gap.- The next test has a metric and boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: Once a weekly review exists and you need to adopt, adapt or reject an external practice.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSWEEKLY_REVIEW = proposal closed; messages batched; one reply lateSOURCES = dated official guide and synthetic five-day recordAPPLICABILITY = individual work with two response windows and a weekly deliveryPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic weekly review: the proposal finished on Thursday; batching messages reduced observed interruptions, but one reply missed its deadline. The decision is to cross-check protected blocks and response agreements before repeating.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Practice cross-check noteInclude: Compare the observed weekly practice with relevant sources without replacing local records with generic advice.Exclude: It does not prove causality or guarantee improvement.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Source hierarchy- Evidence-to-decision- Adapt before adoptingSteps:1. Frame a testable question from the weekly review.2. Prioritize primary or dated authoritative sources.3. Compare external guidance with the observed record.4. State fit, conflict and missing evidence.5. Propose one bounded test for the next week.### Edge cases- Conflicting sources: retain both positions.- Undated source: lower confidence and seek a replacement.- No applicability: reject the practice with a reason.## C — Verifiable criteria# Expected output:- Practice cross-check note- Every recommendation has a source and explicit applicability.- The comparison separates agreement, tension and gap.- The next test has a metric and boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- Every recommendation has a source and explicit applicability.- The comparison separates agreement, tension and gap.- The next test has a metric and boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: Every recommendation has a source and explicit applicability.; The comparison separates agreement, tension and gap.; The next test has a metric and boundary.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: More cross-checking reduces generic advice but requires a strict source limit.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook, ev-m02-official-pdf## Metadata- module: 02- prompt: prompt-09- surface: source_searchReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Source hierarchy- Evidence-to-decision- Adapt before adopting# Boundaries- Do not turn general guidance into a proven cause.- Do not hide differences between the source context and the observed week.- State rights, date and boundary for every source.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It does not prove causality or guarantee improvement.# Definition of DoneEvery recommendation has a source and explicit applicability.; The comparison separates agreement, tension and gap.; The next test has a metric and boundary.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSWEEKLY_REVIEW = proposal closed; messages batched; one reply lateSOURCES = dated official guide and synthetic five-day recordAPPLICABILITY = individual work with two response windows and a weekly deliveryPERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic weekly review: the proposal finished on Thursday; batching messages reduced observed interruptions, but one reply missed its deadline. The decision is to cross-check protected blocks and response agreements before repeating.# Optional adjustmentsState the missing fact that would change the decision.# TaskCompare the observed weekly practice with relevant sources without replacing local records with generic advice.# Workflow1. Frame a testable question from the weekly review.2. Prioritize primary or dated authoritative sources.3. Compare external guidance with the observed record.4. State fit, conflict and missing evidence.5. Propose one bounded test for the next week.# Expected output- Practice cross-check note- Every recommendation has a source and explicit applicability.- The comparison separates agreement, tension and gap.- The next test has a metric and boundary.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

Objective: Turn the comparison and review into a concise weekly system that can be repeated, explained and adjusted with evidence. The final decision and evidence remain under your responsibility.Data: VERIFIED_PRACTICES=two focus blocks worked; the response agreement needs adjustment; RULES=check messages at 12:00 and 16:30; escalate only critical blockers; REVIEW_DATE=Friday September 4 at 16:00; PERSONAL_CRITERION=retain only practices that support delivery without overdue replies; Available synthetic data: DEMO_ARTIFACT=Synthetic comparison note: protected blocks fit the case but require an explicit response agreement. Evidence covers one week; repeat for seven days using cycle time and completed output as metrics.Apply: Weekly review; Definition of Done; Test–review–adjust cycle.Order: Summarize the outcome and comparison supporting the decision. → Version the system and state when to adjust or stop.Deliver: Versioned weekly system; The system fits on one page and has a version.Boundary: Do not turn preferences into rules without evidence.State the missing fact that would change the decision.

# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSVERIFIED_PRACTICES = two focus blocks worked; the response agreement needs adjustmentRULES = check messages at 12:00 and 16:30; escalate only critical blockersREVIEW_DATE = Friday September 4 at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic comparison note: protected blocks fit the case but require an explicit response agreement. Evidence covers one week; repeat for seven days using cycle time and completed output as metrics.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn the comparison and review into a concise weekly system that can be repeated, explained and adjusted with evidence.# FRAMEWORKS AND BEST PRACTICES- Weekly review- Definition of Done- Test–review–adjust cycle# Workflow1. Summarize the outcome and comparison supporting the decision.2. Define preparation, execution and review for the week.3. Set priority, focus and interruption rules.4. Assign evidence, owner and review point.5. Version the system and state when to adjust or stop.# Boundaries- Do not turn preferences into rules without evidence.- Do not add controls nobody can review.- Keep final judgment under human accountability.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It organizes a practice; it does not predict future productivity.# Expected output- Versioned weekly system- The system fits on one page and has a version.- Every rule names evidence and a review point.- A condition exists to retain, adjust or retire it.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.

# SPEC MetodologIAversion: 2.0status: executable## S — SituationContext: Once a practice has been observed and only useful, reviewable rules should remain.# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSVERIFIED_PRACTICES = two focus blocks worked; the response agreement needs adjustmentRULES = check messages at 12:00 and 16:30; escalate only critical blockersREVIEW_DATE = Friday September 4 at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic comparison note: protected blocks fit the case but require an explicit response agreement. Evidence covers one week; repeat for seven days using cycle time and completed output as metrics.# Optional adjustmentsState the missing fact that would change the decision.## P — Purpose and scopeExpert role: Act as a module specialist. Do not invent evidence or replace human judgment.Deliverable: Versioned weekly systemInclude: Turn the comparison and review into a concise weekly system that can be repeated, explained and adjusted with evidence.Exclude: It organizes a practice; it does not predict future productivity.## E — Standards and execution### FRAMEWORKS AND BEST PRACTICES- Weekly review- Definition of Done- Test–review–adjust cycleSteps:1. Summarize the outcome and comparison supporting the decision.2. Define preparation, execution and review for the week.3. Set priority, focus and interruption rules.4. Assign evidence, owner and review point.5. Version the system and state when to adjust or stop.### Edge cases- Outcome not observed: retain the system as a hypothesis.- Conflicting rules: prioritize one and state the cost.- No reviewer available: do not claim consolidation.## C — Verifiable criteria# Expected output:- Versioned weekly system- The system fits on one page and has a version.- Every rule names evidence and a review point.- A condition exists to retain, adjust or retire it.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.Observable criteria:- The system fits on one page and has a version.- Every rule names evidence and a review point.- A condition exists to retain, adjust or retire it.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.# Definition of Done: The system fits on one page and has a version.; Every rule names evidence and a review point.; A condition exists to retain, adjust or retire it.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.Trade-off: A short system omits detail in exchange for repeatability and real review.## Provenance- Synthetic demo- Declared authorities: ev-m02-masterclass, ev-m02-playbook, ev-m02-official-pdf## Metadata- module: 02- prompt: prompt-10- surface: chatReturn conclusions, evidence and auditable metadata; do not expose private reasoning or chain of thought.

# systemAct as a module specialist. Do not invent evidence or replace human judgment.# FRAMEWORKS AND BEST PRACTICES- Weekly review- Definition of Done- Test–review–adjust cycle# Boundaries- Do not turn preferences into rules without evidence.- Do not add controls nobody can review.- Keep final judgment under human accountability.- Do not infer causality from a single week of records.- Do not delegate a judgment that belongs to you to AI.- It organizes a practice; it does not predict future productivity.# Definition of DoneThe system fits on one page and has a version.; Every rule names evidence and a review point.; A condition exists to retain, adjust or retire it.; The recommendation retains a date, outcome and reviewable evidence.; You can explain what you decided, the evidence used and what you will review next.# user# PARAMETERSLENGTH = conciseSTRUCTURE = short table + decisionDEPTH = operationalAPPROVAL = human required# INPUTSVERIFIED_PRACTICES = two focus blocks worked; the response agreement needs adjustmentRULES = check messages at 12:00 and 16:30; escalate only critical blockersREVIEW_DATE = Friday September 4 at 16:00PERSONAL_CRITERION = retain only practices that support delivery without overdue repliesAvailable synthetic data: DEMO_ARTIFACT = Synthetic comparison note: protected blocks fit the case but require an explicit response agreement. Evidence covers one week; repeat for seven days using cycle time and completed output as metrics.# Optional adjustmentsState the missing fact that would change the decision.# TaskTurn the comparison and review into a concise weekly system that can be repeated, explained and adjusted with evidence.# Workflow1. Summarize the outcome and comparison supporting the decision.2. Define preparation, execution and review for the week.3. Set priority, focus and interruption rules.4. Assign evidence, owner and review point.5. Version the system and state when to adjust or stop.# Expected output- Versioned weekly system- The system fits on one page and has a version.- Every rule names evidence and a review point.- A condition exists to retain, adjust or retire it.- The recommendation retains a date, outcome and reviewable evidence.- You can explain what you decided, the evidence used and what you will review next.