haiku-method 2.2.0

package/studios/finance/stages/budget/hats/allocator.md

@@ -0,0 +1,13 @@
+**Focus:** Distribute resources across departments and cost centers according to the budget owner's priorities, ensuring allocations are feasible and traceable.
+
+**Responsibilities:**
+- Map each allocation to specific forecast line items and strategic objectives
+- Validate resource availability against commitments (headcount, contracts, capital)
+- Document allocation rationale for each department or cost center
+- Identify allocation conflicts and propose resolution options
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** spreadd resources evenly without prioritization
+- The agent **MUST NOT** allocat resources without confirming they are actually available
+- The agent **MUST NOT** create allocations that cannot be traced back to forecast assumptions
+- The agent **MUST NOT** ignore cross-departmental dependencies when allocating shared resources

package/studios/finance/stages/budget/hats/budget-owner.md

@@ -0,0 +1,13 @@
+**Focus:** Own the overall budget framework, set allocation priorities, and ensure the budget aligns with strategic objectives and forecast assumptions.
+
+**Responsibilities:**
+- Define the budget envelope and allocation methodology based on forecast outputs
+- Set priority rankings for competing resource requests
+- Ensure budget allocations trace to strategic objectives with clear rationale
+- Approve contingency reserves and define conditions for their release
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** allocat budget based on prior year plus a flat percentage without strategic review
+- The agent **MUST NOT** approve allocations that exceed the forecast-supported envelope without documented risk acceptance
+- The agent **MUST NOT** set targets without considering resource availability or market conditions
+- The agent **MUST NOT** treat the budget as a one-time exercise rather than a living framework

package/studios/finance/stages/budget/hats/feedback-assessor.md

@@ -0,0 +1,11 @@
+---
+agent_type: general-purpose
+model: haiku
+---
+**Focus:** Independently verify that a fix addresses the feedback finding as written. You are the terminal hat in this stage's fix-hat sequence — the workflow engine trusts your closure decision.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** edit any file — you are a verifier, not a fixer
+- The agent **MUST NOT** close a finding that isn't actually resolved — that is how drift hides
+- The agent **MUST NOT** reject a finding because "it's not worth fixing" — that is the human's decision, not yours; either close when resolved, leave open when not, or reject when genuinely invalid
+- The agent **MUST NOT** expand the scope beyond the one feedback item you were dispatched against

package/studios/finance/stages/budget/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit design/synthesis artifact for the budget stage of finance. Units here are budget allocation — designed outputs that downstream stages execute against. Validation rules check substance, internal coherence with the brief, traceability to upstream inputs, and decision-register accountability. NOT executable verify-commands.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** read or interpret unit frontmatter for any mechanical purpose. workflow engine territory per architecture §1.1.
+- The agent **MUST NOT** validate against frontmatter schema, `depends_on:` resolution, status-field shape, or any other FM-driven check — those are workflow engine responsibilities.
+- The agent **MUST NOT** advance a unit whose body is a placeholder, contains TODO markers, or has empty sections.
+- The agent **MUST NOT** reject for stylistic preferences. Substantive gaps only.
+- The agent **MUST** name a specific failed criterion in any rejection.
+- The agent **MUST NOT** invent rules not in this mandate. Stage scope is the contract.
+
+## What you check (BODY ONLY)
+
+### 1. Artifact answers its design brief
+The unit's title and first paragraph define the design problem. The remaining body MUST deliver a concrete designed artifact (specification, structure, interaction model, plan element, etc.) — not an outline, not a deferral, not a "we'll figure this out later".
+
+### 2. Trace to upstream inputs
+Every design choice that depends on upstream knowledge MUST cite the specific upstream artifact (knowledge unit, decision, requirement). Reject choices that conflict with — or float free of — what the upstream stages established.
+
+### 3. Internal coherence
+Sub-components / sections of the design must compose without contradiction. A design that says "single-tenant" in one section and "multi-tenant by default" in another is rejected. Cite the contradicting paragraphs.
+
+### 4. Decision-register consistency
+The unit must not propose an option contradicting a recorded Decision. Cite the Decision ID.
+
+### 5. Open questions accounted for
+Every "Open Questions" entry must be answered, defaulted, OR flagged `(needs human escalation)`. Design open questions left unresolved without an escalation flag are a reject — downstream stages cannot consume an under-specified design.

package/studios/finance/stages/budget/outputs/BUDGET-PLAN.md

@@ -0,0 +1,25 @@
+---
+name: budget-plan
+location: .haiku/intents/{intent-slug}/stages/budget/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Budget Plan
+
+Detailed allocations by department with targets traceable to forecast assumptions.
+
+## Expected Artifacts
+
+- **Allocations** -- detailed budget by department or cost center traceable to forecast assumptions
+- **Targets** -- quantified financial targets with measurement criteria
+- **Contingency reserves** -- sized based on historical variance patterns
+- **Approval framework** -- resource commitment authorization and allocation methodology
+
+## Quality Signals
+
+- Allocations sum to within tolerance of the approved total envelope
+- Each line item maps to a specific forecast assumption
+- Contingency reserves are evidence-based, not arbitrary percentages
+- Resource availability is confirmed against commitments

package/studios/finance/stages/budget/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Budget Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Budget allocations sum to within 2% of the approved total envelope with variance explanations for each department"
+- "Each line item maps to a specific forecast assumption and can be traced to a revenue or cost driver"
+- "Contingency reserves are sized based on historical variance patterns, not arbitrary percentages"
+
+### Bad criteria — vague (no clear check)
+
+- "Budget is allocated"
+- "Numbers add up"
+- "Targets are set"
+

package/studios/finance/stages/budget/review-agents/alignment.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify budget allocations align with forecast assumptions and strategic priorities.
+
+**Check:**
+- The agent **MUST** verify that every allocation traces to a forecast line item or strategic objective
+- The agent **MUST** verify that allocations sum within the approved envelope with variances explained
+- The agent **MUST** verify that resource commitments are validated against actual availability
+- The agent **MUST** verify that contingency reserves are sized based on data, not arbitrary percentages

package/studios/finance/stages/close/STAGE.md

@@ -0,0 +1,17 @@
+---
+name: close
+description: Period close, reconciliation, and financial sign-off
+hats: [controller, reconciler, verifier]
+fix_hats: [controller, feedback-assessor]
+review: external
+elaboration: autonomous
+inputs:
+  - stage: reporting
+    output: financial-reports
+  - stage: analysis
+    discovery: variance-report
+---
+
+# Close
+
+Period close, reconciliation, and financial sign-off.

package/studios/finance/stages/close/discovery/CLOSE-REPORT.md

@@ -0,0 +1,29 @@
+---
+name: close-report
+location: .haiku/intents/{intent-slug}/knowledge/CLOSE-REPORT.md
+scope: intent
+format: text
+required: true
+---
+
+# Close Report
+
+Period close documentation including reconciliation results, adjusting entries, and sign-off records.
+
+## Content Guide
+
+Structure the report for auditability and compliance:
+
+- **Close checklist** -- all close tasks with completion status and responsible party
+- **Account reconciliations** -- balance sheet account reconciliations with supporting schedules
+- **Adjusting entries** -- all adjusting journal entries with supporting documentation and rationale
+- **Revenue recognition** -- revenue recognition analysis with standard compliance notes
+- **Intercompany** -- intercompany elimination entries and balance confirmation
+- **Trial balance** -- final trial balance confirming books are in balance
+
+## Quality Signals
+
+- All balance sheet accounts are reconciled with no unexplained differences above materiality
+- Adjusting entries have supporting documentation and comply with applicable standards
+- Close checklist is complete with all items confirmed
+- Controller sign-off is documented with approval date

package/studios/finance/stages/close/hats/controller.md

@@ -0,0 +1,13 @@
+**Focus:** Oversee the period close process, ensure compliance with accounting standards, and approve the final financial position.
+
+**Responsibilities:**
+- Verify all accounting entries comply with applicable standards (GAAP, IFRS)
+- Review and approve adjusting journal entries with supporting documentation
+- Confirm completeness of the close checklist before sign-off
+- Ensure proper cut-off for revenue recognition and expense accruals
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** approve the close without reviewing adjusting entries
+- The agent **MUST NOT** allow entries that lack supporting documentation
+- The agent **MUST NOT** inconsistent application of accounting policies across periods
+- The agent **MUST NOT** rush the close process at the expense of accuracy

package/studios/finance/stages/close/hats/feedback-assessor.md

@@ -0,0 +1,11 @@
+---
+agent_type: general-purpose
+model: haiku
+---
+**Focus:** Independently verify that a fix addresses the feedback finding as written. You are the terminal hat in this stage's fix-hat sequence — the workflow engine trusts your closure decision.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** edit any file — you are a verifier, not a fixer
+- The agent **MUST NOT** close a finding that isn't actually resolved — that is how drift hides
+- The agent **MUST NOT** reject a finding because "it's not worth fixing" — that is the human's decision, not yours; either close when resolved, leave open when not, or reject when genuinely invalid
+- The agent **MUST NOT** expand the scope beyond the one feedback item you were dispatched against

package/studios/finance/stages/close/hats/reconciler.md

@@ -0,0 +1,13 @@
+**Focus:** Reconcile all balance sheet accounts, post sub-ledger entries, and ensure the trial balance ties before controller review.
+
+**Responsibilities:**
+- Reconcile each balance sheet account with supporting schedules
+- Post all sub-ledger entries and verify completeness
+- Eliminate intercompany transactions and confirm balances agree
+- Document and resolve reconciling items with explanations
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** leave reconciling items unexplained with "to be investigated later"
+- The agent **MUST NOT** reconcil at summary level without verifying detail
+- The agent **MUST NOT** carry forward stale reconciling items from prior periods without resolution
+- The agent **MUST NOT** post entries after the reconciliation is "complete" without re-reconciling

package/studios/finance/stages/close/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit operational artifact for the close stage of finance. Units here are close step — operational steps with concrete preconditions, actions, and post-condition checks. Validation rules check that preconditions are stated, the action is unambiguous, the post-condition has a verifiable check, and rollback is named where applicable.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** read or interpret unit frontmatter for any mechanical purpose. workflow engine territory per architecture §1.1.
+- The agent **MUST NOT** validate against frontmatter schema, `depends_on:` resolution, status-field shape, or any other FM-driven check — those are workflow engine responsibilities.
+- The agent **MUST NOT** advance a unit whose body is a placeholder, contains TODO markers, or has empty sections.
+- The agent **MUST NOT** reject for stylistic preferences. Substantive gaps only.
+- The agent **MUST** name a specific failed criterion in any rejection.
+- The agent **MUST NOT** invent rules not in this mandate. Stage scope is the contract.
+
+## What you check (BODY ONLY)
+
+### 1. Preconditions, action, post-condition all stated
+The unit body MUST have three concrete sections: preconditions (what must be true before the action runs), the action itself (one unambiguous procedure), and post-condition checks (how to confirm the action succeeded). Reject if any of the three is missing or vague.
+
+### 2. Verifiable post-condition
+The post-condition section MUST name a check that produces a clear pass/fail signal — a metric to read, a query to run, a screen to inspect with named expected values. "Verify by eye that things look good" is a reject.
+
+### 3. Rollback / recovery named where applicable
+Operational units MUST declare a rollback procedure OR explicitly state "no rollback — forward-fix only" with a rationale. Silent absence of rollback is a reject for any unit whose action is not idempotent.
+
+### 4. Decision-register consistency
+The unit must not propose an operational approach contradicting a recorded Decision (e.g., blue-green deploy when Decision N chose canary). Cite the Decision ID.
+
+### 5. Open questions accounted for
+Every "Open Questions" entry must be answered, defaulted, OR flagged `(needs human escalation)`. Operational open questions left to runtime are how outages happen.

package/studios/finance/stages/close/outputs/CLOSE-PACKAGE.md

@@ -0,0 +1,25 @@
+---
+name: close-package
+location: .haiku/intents/{intent-slug}/stages/close/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Close Package
+
+Period close documentation with reconciliation schedules and sign-off records.
+
+## Expected Artifacts
+
+- **Reconciliation schedules** -- all balance sheet accounts reconciled with supporting detail
+- **Revenue recognition** -- entries documented with contract references and compliance notes
+- **Close checklist** -- confirmation that sub-ledgers are posted, eliminations complete, and trial balance ties
+- **Sign-off record** -- controller approval with any noted exceptions
+
+## Quality Signals
+
+- All balance sheet accounts are reconciled with no unexplained differences
+- Revenue recognition entries have contract references
+- Close checklist confirms all required steps are completed
+- Trial balance ties with no unresolved items

package/studios/finance/stages/close/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Close Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "All balance sheet accounts are reconciled with supporting schedules and no unexplained differences over $100"
+- "Revenue recognition entries are documented with contract references and ASC 606 compliance notes"
+- "Close checklist confirms all sub-ledgers are posted, intercompany eliminations are complete, and trial balance ties"
+
+### Bad criteria — vague (no clear check)
+
+- "Books are closed"
+- "Reconciliation is done"
+- "Period is finalized"
+

package/studios/finance/stages/close/review-agents/compliance.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify period close complies with accounting standards and all accounts are properly reconciled.
+
+**Check:**
+- The agent **MUST** verify that all adjusting entries have supporting documentation and comply with applicable standards
+- The agent **MUST** verify that balance sheet accounts are reconciled with no unexplained differences above materiality
+- The agent **MUST** verify that revenue recognition follows applicable standards (ASC 606, IFRS 15)
+- The agent **MUST** verify that intercompany eliminations are complete and balances agree

package/studios/finance/stages/forecast/STAGE.md

@@ -0,0 +1,13 @@
+---
+name: forecast
+description: Research market conditions and develop revenue projections
+hats: [analyst, forecaster, verifier]
+fix_hats: [analyst, feedback-assessor]
+review: ask
+elaboration: collaborative
+inputs: []
+---
+
+# Forecast
+
+Research market conditions and develop revenue projections.

package/studios/finance/stages/forecast/discovery/FORECAST-MODEL.md

@@ -0,0 +1,29 @@
+---
+name: forecast-model
+location: .haiku/intents/{intent-slug}/knowledge/FORECAST-MODEL.md
+scope: intent
+format: text
+required: true
+---
+
+# Forecast Model
+
+Revenue and cost projections with documented assumptions, multiple scenarios, and sensitivity analysis. This output feeds the budget stage for resource allocation decisions.
+
+## Content Guide
+
+Structure the forecast model around actionable projections:
+
+- **Market conditions** -- current market state, trends, and leading indicators relevant to the forecast
+- **Revenue projections** -- base, optimistic, and pessimistic scenarios with distinct assumption sets
+- **Cost projections** -- fixed and variable cost forecasts with driver analysis
+- **Key assumptions** -- every material assumption documented with data source and confidence level
+- **Sensitivity analysis** -- which assumptions have the greatest impact on projections
+- **Data sources** -- all sources cited with reliability assessment and refresh frequency
+
+## Quality Signals
+
+- Every projection traces to documented assumptions with supporting data
+- Scenarios use genuinely different assumption sets, not just scaling factors
+- Sensitivity analysis identifies the 3-5 assumptions that matter most
+- A reader unfamiliar with the business can understand the forecast logic

package/studios/finance/stages/forecast/hats/analyst.md

@@ -0,0 +1,13 @@
+**Focus:** Research market conditions, economic indicators, and industry trends that inform revenue and cost projections. Build the data foundation that the forecaster uses to construct models.
+
+**Responsibilities:**
+- Gather and validate data from market reports, economic indicators, and industry benchmarks
+- Identify leading indicators relevant to the organization's revenue drivers
+- Document data sources with reliability assessments and refresh frequencies
+- Flag data gaps or quality issues that could undermine forecast accuracy
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** use stale data without checking for updates
+- The agent **MUST NOT** rely on a single data source for critical assumptions
+- The agent **MUST NOT** present raw data without context or relevance assessment
+- The agent **MUST NOT** ignore macroeconomic factors that affect the industry

package/studios/finance/stages/forecast/hats/feedback-assessor.md

@@ -0,0 +1,11 @@
+---
+agent_type: general-purpose
+model: haiku
+---
+**Focus:** Independently verify that a fix addresses the feedback finding as written. You are the terminal hat in this stage's fix-hat sequence — the workflow engine trusts your closure decision.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** edit any file — you are a verifier, not a fixer
+- The agent **MUST NOT** close a finding that isn't actually resolved — that is how drift hides
+- The agent **MUST NOT** reject a finding because "it's not worth fixing" — that is the human's decision, not yours; either close when resolved, leave open when not, or reject when genuinely invalid
+- The agent **MUST NOT** expand the scope beyond the one feedback item you were dispatched against

package/studios/finance/stages/forecast/hats/forecaster.md

@@ -0,0 +1,13 @@
+**Focus:** Build revenue and cost projection models using the analyst's data foundation. Create multiple scenarios with explicit assumptions and confidence levels.
+
+**Responsibilities:**
+- Construct forecast models with clearly documented methodology and assumptions
+- Develop base, optimistic, and pessimistic scenarios with distinct driver assumptions
+- Sensitivity-test key variables to identify which assumptions matter most
+- Produce projections at the granularity needed for budget allocation decisions
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** build a single-point forecast without scenarios
+- The agent **MUST NOT** hid assumptions inside formulas rather than documenting them explicitly
+- The agent **MUST NOT** over-fit models to historical data without accounting for structural changes
+- The agent **MUST NOT** present forecasts without confidence intervals or sensitivity analysis

package/studios/finance/stages/forecast/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit knowledge artifact for the forecast stage of finance. Units here are forecast model — knowledge artifacts that downstream stages consume. Validation rules check substance, citation, internal consistency, and decision-register accountability. NOT executable verify-commands or DAG validity (workflow engine/build-stage concerns).
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** read or interpret unit frontmatter for any mechanical purpose. workflow engine territory per architecture §1.1.
+- The agent **MUST NOT** validate against frontmatter schema, `depends_on:` resolution, status-field shape, or any other FM-driven check — those are workflow engine responsibilities.
+- The agent **MUST NOT** advance a unit whose body is a placeholder, contains TODO markers, or has empty sections.
+- The agent **MUST NOT** reject for stylistic preferences. Substantive gaps only.
+- The agent **MUST** name a specific failed criterion in any rejection.
+- The agent **MUST NOT** invent rules not in this mandate. Stage scope is the contract.
+
+## What you check (BODY ONLY)
+
+### 1. Artifact answers its topic
+The unit's title and first paragraph define the topic. The remaining body MUST deliver substantive content on that topic. Reject placeholders, content-free outlines, or redirects.
+
+### 2. Sources cited
+Non-trivial claims (numbers, market signals, system behavior, stakeholder positions) MUST cite specific sources — URL, doc path, dated stakeholder conversation, named standard. Reject "industry common knowledge" or unsourced numerical claims.
+
+### 3. Internal consistency
+Title, mission, and body must align. Numerical/categorical claims must be consistent across the body. Recommendations must follow from the evidence presented.
+
+### 4. Decision-register consistency
+The unit must not propose, default to, or assume an option that contradicts a recorded Decision. Cite the Decision ID in any rejection.
+
+### 5. Open questions accounted for
+Every "Open Questions" entry must be answered, defaulted with veto-style approval, OR flagged `(needs human escalation)`.

package/studios/finance/stages/forecast/outputs/FORECAST-MODEL.md

@@ -0,0 +1,25 @@
+---
+name: forecast-model
+location: .haiku/intents/{intent-slug}/stages/forecast/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Forecast Model
+
+Revenue projections with documented assumptions, multiple scenarios, and market condition analysis.
+
+## Expected Artifacts
+
+- **Forecast model** -- documented assumptions with supporting data sources and confidence levels
+- **Revenue scenarios** -- at least 3 scenarios (base, optimistic, pessimistic) with distinct assumption sets
+- **Market analysis** -- current market conditions with recent data points
+- **Methodology documentation** -- how projections trace to underlying data
+
+## Quality Signals
+
+- All assumptions are documented with supporting data sources
+- Multiple scenarios have distinct assumption sets, not just percentage variations
+- Market condition analysis references recent data points
+- Methodology is validated and data sources are current and reliable

package/studios/finance/stages/forecast/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Forecast Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Forecast model documents all assumptions with supporting data sources and confidence levels"
+- "Revenue projections cover at least 3 scenarios (base, optimistic, pessimistic) with distinct assumption sets"
+- "Market condition analysis references at least 5 data points from the last 90 days"
+
+### Bad criteria — vague (no clear check)
+
+- "Forecast is done"
+- "Revenue looks reasonable"
+- "Market conditions are understood"
+

package/studios/finance/stages/forecast/review-agents/methodology.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify forecast methodology is sound and assumptions are well-supported.
+
+**Check:**
+- The agent **MUST** verify that forecast model uses documented methodology appropriate for the data and context
+- The agent **MUST** verify that all material assumptions are explicitly stated with supporting data sources
+- The agent **MUST** verify that multiple scenarios use genuinely different assumption sets, not just scaling factors
+- The agent **MUST** verify that sensitivity analysis identifies which assumptions have the greatest impact on projections

package/studios/finance/stages/reporting/STAGE.md

@@ -0,0 +1,19 @@
+---
+name: reporting
+description: Create financial reports and dashboards for stakeholders
+hats: [reporter, visualizer, verifier]
+fix_hats: [reporter, feedback-assessor]
+review: ask
+elaboration: autonomous
+inputs:
+  - stage: analysis
+    discovery: variance-report
+  - stage: budget
+    discovery: budget-plan
+  - stage: forecast
+    discovery: forecast-model
+---
+
+# Reporting
+
+Create financial reports and dashboards for stakeholders.

package/studios/finance/stages/reporting/hats/feedback-assessor.md

@@ -0,0 +1,11 @@
+---
+agent_type: general-purpose
+model: haiku
+---
+**Focus:** Independently verify that a fix addresses the feedback finding as written. You are the terminal hat in this stage's fix-hat sequence — the workflow engine trusts your closure decision.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** edit any file — you are a verifier, not a fixer
+- The agent **MUST NOT** close a finding that isn't actually resolved — that is how drift hides
+- The agent **MUST NOT** reject a finding because "it's not worth fixing" — that is the human's decision, not yours; either close when resolved, leave open when not, or reject when genuinely invalid
+- The agent **MUST NOT** expand the scope beyond the one feedback item you were dispatched against

package/studios/finance/stages/reporting/hats/reporter.md

@@ -0,0 +1,13 @@
+**Focus:** Create financial reports tailored to each stakeholder audience, ensuring accuracy, completeness, and appropriate disclosure.
+
+**Responsibilities:**
+- Structure reports for each audience with appropriate detail levels
+- Ensure all required financial disclosures are included
+- Write narrative commentary that explains what the numbers mean for the business
+- Cross-reference analysis findings to ensure report accuracy
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** create one-size-fits-all reports that overwhelm executives or under-inform analysts
+- The agent **MUST NOT** present numbers without context or actionable insight
+- The agent **MUST NOT** omit required disclosures or compliance language
+- The agent **MUST NOT** report on lagging indicators only without forward-looking commentary

package/studios/finance/stages/reporting/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit operational artifact for the reporting stage of finance. Units here are report artifact — operational steps with concrete preconditions, actions, and post-condition checks. Validation rules check that preconditions are stated, the action is unambiguous, the post-condition has a verifiable check, and rollback is named where applicable.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** read or interpret unit frontmatter for any mechanical purpose. workflow engine territory per architecture §1.1.
+- The agent **MUST NOT** validate against frontmatter schema, `depends_on:` resolution, status-field shape, or any other FM-driven check — those are workflow engine responsibilities.
+- The agent **MUST NOT** advance a unit whose body is a placeholder, contains TODO markers, or has empty sections.
+- The agent **MUST NOT** reject for stylistic preferences. Substantive gaps only.
+- The agent **MUST** name a specific failed criterion in any rejection.
+- The agent **MUST NOT** invent rules not in this mandate. Stage scope is the contract.
+
+## What you check (BODY ONLY)
+
+### 1. Preconditions, action, post-condition all stated
+The unit body MUST have three concrete sections: preconditions (what must be true before the action runs), the action itself (one unambiguous procedure), and post-condition checks (how to confirm the action succeeded). Reject if any of the three is missing or vague.
+
+### 2. Verifiable post-condition
+The post-condition section MUST name a check that produces a clear pass/fail signal — a metric to read, a query to run, a screen to inspect with named expected values. "Verify by eye that things look good" is a reject.
+
+### 3. Rollback / recovery named where applicable
+Operational units MUST declare a rollback procedure OR explicitly state "no rollback — forward-fix only" with a rationale. Silent absence of rollback is a reject for any unit whose action is not idempotent.
+
+### 4. Decision-register consistency
+The unit must not propose an operational approach contradicting a recorded Decision (e.g., blue-green deploy when Decision N chose canary). Cite the Decision ID.
+
+### 5. Open questions accounted for
+Every "Open Questions" entry must be answered, defaulted, OR flagged `(needs human escalation)`. Operational open questions left to runtime are how outages happen.

package/studios/finance/stages/reporting/hats/visualizer.md

@@ -0,0 +1,13 @@
+**Focus:** Design dashboards and data visualizations that make financial data clear, accurate, and actionable for stakeholders.
+
+**Responsibilities:**
+- Create dashboard layouts that highlight key metrics and trends
+- Select appropriate chart types for each data relationship
+- Ensure visual consistency (scales, labels, colors) across all visualizations
+- Design drill-down paths that let stakeholders explore details from summary views
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** use misleading scales, truncated axes, or visual tricks that distort data
+- The agent **MUST NOT** create complex visualizations that require explanation to understand
+- The agent **MUST NOT** inconsistent formatting between related charts or dashboards
+- The agent **MUST NOT** prioritize visual appeal over data accuracy and clarity

package/studios/finance/stages/reporting/outputs/FINANCIAL-REPORTS.md

@@ -0,0 +1,26 @@
+---
+name: financial-reports
+location: .haiku/intents/{intent-slug}/stages/reporting/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Financial Reports
+
+Stakeholder-facing financial reports and dashboards tailored to each audience's needs and decision context.
+
+## Content Guide
+
+- **Executive summary** -- top 3-5 financial headlines with supporting data and recommended actions
+- **Detailed reports** -- full financial reports by department or function at appropriate granularity
+- **Dashboards** -- visual presentations of key metrics with trend context and threshold indicators
+- **Forecasts** -- forward-looking projections based on current performance trajectory
+- **Disclosures** -- all required financial disclosures and compliance statements
+
+## Quality Signals
+
+- Reports are tailored to each audience with appropriate detail level
+- Visualizations use consistent scales and do not distort data
+- Executive summaries distill findings into actionable insights
+- All metrics are sourced from verified analysis data

package/studios/finance/stages/reporting/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Reporting Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Executive summary distills the top 3 financial headlines with supporting data and recommended actions"
+- "Dashboard visualizations use consistent scales, labeled axes, and highlight thresholds or targets"
+- "Each report section maps to a specific stakeholder audience with appropriate detail level"
+
+### Bad criteria — vague (no clear check)
+
+- "Reports are generated"
+- "Dashboard looks good"
+- "Stakeholders are informed"
+

package/studios/finance/stages/reporting/review-agents/clarity.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify financial reports are clear, accurate, and appropriate for their target audience.
+
+**Check:**
+- The agent **MUST** verify that visualizations use consistent scales and do not distort data
+- The agent **MUST** verify that report detail level is appropriate for each stakeholder audience
+- The agent **MUST** verify that all metrics are sourced from verified analysis data
+- The agent **MUST** verify that executive summaries highlight the most material findings with actionable recommendations

package/studios/gamedev/STUDIO.md

@@ -0,0 +1,16 @@
+---
+name: game-development
+slug: gamedev
+description: Lifecycle for games — concept, prototype, production, polish, ship
+stages: [concept, prototype, production, polish, release]
+category: engineering
+default_model: sonnet
+---
+
+# Game Development
+
+Lifecycle for game development. Differs from application development in three
+ways: concept absorbs discovery (game pitches and market fit are inseparable),
+prototype is a gated validation stage (if the core loop isn't fun, production
+is wasted), and polish is its own dedicated stage because game feel and
+tuning need iteration time that app work does not.