haiku-method 2.2.0

package/studios/executive-strategy/stages/landscape/STAGE.md

@@ -0,0 +1,13 @@
+---
+name: landscape
+description: Analyze market conditions, competitive intelligence, and strategic context
+hats: [strategist, analyst, verifier]
+fix_hats: [strategist, feedback-assessor]
+review: auto
+elaboration: autonomous
+inputs: []
+---
+
+# Landscape
+
+Analyze market conditions, competitive intelligence, and strategic context.

package/studios/executive-strategy/stages/landscape/discovery/LANDSCAPE-ANALYSIS.md

@@ -0,0 +1,29 @@
+---
+name: landscape-analysis
+location: .haiku/intents/{intent-slug}/knowledge/LANDSCAPE-ANALYSIS.md
+scope: intent
+format: text
+required: true
+---
+
+# Landscape Analysis
+
+Market conditions, competitive intelligence, and organizational context that frames the strategic decision.
+
+## Content Guide
+
+Structure the analysis to enable informed option generation:
+
+- **Strategic question** -- the decision to be made with scope and boundaries
+- **Market trends** -- key trends with evidence-based impact assessment
+- **Competitive landscape** -- competitor positions, strategies, and likely next moves
+- **Internal capabilities** -- organizational strengths, weaknesses, and resource position
+- **Key uncertainties** -- the factors most likely to shape outcomes, with current assessment
+- **SWOT synthesis** -- integrated view connecting external environment to internal capabilities
+
+## Quality Signals
+
+- Trends are supported by current data from authoritative sources
+- Competitive intelligence reflects actual positions, not assumptions
+- Internal assessment is honest about weaknesses, not just strengths
+- Analysis is synthesized into a coherent view, not a data dump

package/studios/executive-strategy/stages/landscape/hats/analyst.md

@@ -0,0 +1,13 @@
+**Focus:** Gather and validate data on market conditions, competitive positions, and industry trends that inform strategic decisions.
+
+**Responsibilities:**
+- Research market size, growth rates, and trend data from authoritative sources
+- Map competitor strategies, market positions, and recent strategic moves
+- Validate data quality and document source reliability
+- Identify data gaps and their potential impact on analysis quality
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** rely on outdated market reports without checking for recent changes
+- The agent **MUST NOT** present data without assessing source credibility
+- The agent **MUST NOT** focus solely on direct competitors while ignoring adjacent market threats
+- The agent **MUST NOT** collect data without a clear framework for what matters and why

package/studios/executive-strategy/stages/landscape/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/executive-strategy/stages/landscape/hats/strategist.md

@@ -0,0 +1,13 @@
+**Focus:** Frame the strategic question, define the analysis scope, and synthesize market and competitive intelligence into a coherent landscape view.
+
+**Responsibilities:**
+- Frame the strategic decision with clear boundaries and success criteria
+- Synthesize market trends, competitive dynamics, and internal capabilities into a unified view
+- Identify the key forces and uncertainties that shape the decision context
+- Structure the landscape analysis for consumption by the options generation stage
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** fram the question too narrowly, precluding creative strategic options
+- The agent **MUST NOT** present data without a synthesized point of view on what it means
+- The agent **MUST NOT** ignore internal capabilities and constraints when mapping the external landscape
+- The agent **MUST NOT** treat landscape analysis as a one-time snapshot rather than a dynamic picture

package/studios/executive-strategy/stages/landscape/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit knowledge artifact for the landscape stage of executive-strategy. Units here are landscape analysis — 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/executive-strategy/stages/landscape/outputs/LANDSCAPE-ANALYSIS.md

@@ -0,0 +1,25 @@
+---
+name: landscape-analysis
+location: .haiku/intents/{intent-slug}/stages/landscape/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Landscape Analysis
+
+Market conditions, competitive intelligence, and strategic context assessment.
+
+## Expected Artifacts
+
+- **Market trends** -- identified trends with evidence-based impact assessment
+- **Competitive positions** -- each competitor's strategic position, recent moves, and likely next actions
+- **SWOT analysis** -- each element connected to specific, verifiable facts
+- **Decision context** -- strategic framing for the decision at hand
+
+## Quality Signals
+
+- Market trends are evidence-based, not speculative
+- Competitive intelligence maps specific positions and recent actions
+- SWOT elements connect to verifiable facts, not generic observations
+- Data sources are validated and analytical framework is sound

package/studios/executive-strategy/stages/landscape/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Landscape Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Market analysis identifies at least 5 trends with evidence-based assessment of their impact on the organization"
+- "Competitive intelligence maps each competitor's strategic position, recent moves, and likely next actions"
+- "SWOT analysis connects each element to specific, verifiable facts rather than generic observations"
+
+### Bad criteria — vague (no clear check)
+
+- "Market is understood"
+- "Competition is analyzed"
+- "Landscape is mapped"
+

package/studios/executive-strategy/stages/landscape/review-agents/rigor.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify landscape analysis is evidence-based and covers all relevant dimensions.
+
+**Check:**
+- The agent **MUST** verify that market trends are supported by current data sources, not outdated reports
+- The agent **MUST** verify that competitive intelligence reflects actual competitor positions, not assumptions
+- The agent **MUST** verify that internal capability assessment is honest, not self-congratulatory
+- The agent **MUST** verify that analysis framework is appropriate for the strategic question being addressed

package/studios/executive-strategy/stages/options/STAGE.md

@@ -0,0 +1,15 @@
+---
+name: options
+description: Generate and structure strategic options
+hats: [ideator, modeler, verifier]
+fix_hats: [ideator, feedback-assessor]
+review: ask
+elaboration: collaborative
+inputs:
+  - stage: landscape
+    discovery: landscape-analysis
+---
+
+# Options
+
+Generate and structure strategic options.

package/studios/executive-strategy/stages/options/discovery/OPTIONS-MATRIX.md

@@ -0,0 +1,29 @@
+---
+name: options-matrix
+location: .haiku/intents/{intent-slug}/knowledge/OPTIONS-MATRIX.md
+scope: intent
+format: text
+required: true
+---
+
+# Options Matrix
+
+Strategic alternatives with financial models, resource requirements, and risk profiles.
+
+## Content Guide
+
+Structure the matrix for comparative evaluation:
+
+- **Option summaries** -- concise description of each strategic alternative with theory of change
+- **Financial models** -- investment required, projected returns, and break-even for each option
+- **Resource requirements** -- capital, talent, time, and organizational capacity needed
+- **Risk profiles** -- key risks for each option with probability and impact
+- **Key assumptions** -- material assumptions for each option with sensitivity indicators
+- **Comparison framework** -- dimensions for evaluating options in the next stage
+
+## Quality Signals
+
+- Options are genuinely distinct, not variations of the same approach
+- Financial models use consistent assumptions for fair comparison
+- Risk profiles span a range from conservative to aggressive
+- Each option has a credible theory of change with stated assumptions

package/studios/executive-strategy/stages/options/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/executive-strategy/stages/options/hats/ideator.md

@@ -0,0 +1,13 @@
+**Focus:** Generate genuinely differentiated strategic options that span a range of risk/reward profiles.
+
+**Responsibilities:**
+- Generate multiple strategic options that address the landscape findings
+- Ensure options are genuinely distinct, not variations of the same approach
+- Include both conventional and unconventional options to expand the decision space
+- Articulate the value proposition and theory of change for each option
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** generate only safe, incremental options without any bold alternatives
+- The agent **MUST NOT** create options that are really just implementation variants of the same strategy
+- The agent **MUST NOT** anchor on a preferred option and generating others as strawmen
+- The agent **MUST NOT** ignore options that challenge current organizational assumptions

package/studios/executive-strategy/stages/options/hats/modeler.md

@@ -0,0 +1,13 @@
+**Focus:** Build financial and operational models for each strategic option to quantify investment, returns, and resource requirements.
+
+**Responsibilities:**
+- Construct financial models with explicit assumptions for each option
+- Quantify investment required, projected returns, and break-even conditions
+- Model resource requirements (capital, talent, time) for each option
+- Identify the key assumptions that most affect each option's viability
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** build overly complex models that obscure key drivers
+- The agent **MUST NOT** present single-point projections without sensitivity ranges
+- The agent **MUST NOT** use inconsistent assumptions across options, making comparison unfair
+- The agent **MUST NOT** model only financial outcomes without considering operational feasibility

package/studios/executive-strategy/stages/options/hats/verifier.md

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit design/synthesis artifact for the options stage of executive-strategy. Units here are option model — 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/executive-strategy/stages/options/outputs/OPTIONS-MATRIX.md

@@ -0,0 +1,25 @@
+---
+name: options-matrix
+location: .haiku/intents/{intent-slug}/stages/options/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Options Matrix
+
+Distinct strategic alternatives with financial models, resource requirements, and risk profiles.
+
+## Expected Artifacts
+
+- **Strategic options** -- at least 3 distinct paths with differentiated value propositions
+- **Financial models** -- stated assumptions, investment required, and projected returns per option
+- **Risk profiles** -- risk/reward assessment spanning a range of profiles
+- **Resource requirements** -- what each option demands in terms of capital, people, and time
+
+## Quality Signals
+
+- Options are genuinely differentiated, not variations of one approach
+- Financial assumptions are stated and quantified uncertainties are identified
+- Options span a range of risk/reward profiles
+- Each option has enough detail for meaningful evaluation

package/studios/executive-strategy/stages/options/phases/ELABORATION.md

@@ -0,0 +1,16 @@
+# Options Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Options matrix includes at least 3 distinct strategic paths with clearly differentiated value propositions"
+- "Each option has a financial model with stated assumptions, investment required, and projected returns"
+- "Options span a range of risk/reward profiles — not all variations of the same conservative approach"
+
+### Bad criteria — vague (no clear check)
+
+- "Options are generated"
+- "Models are built"
+- "Alternatives are listed"
+

package/studios/executive-strategy/stages/options/review-agents/differentiation.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify strategic options are genuinely distinct and cover an adequate range of risk/reward profiles.
+
+**Check:**
+- The agent **MUST** verify that options represent genuinely different strategic directions, not variations of one approach
+- The agent **MUST** verify that financial models use consistent assumptions for fair comparison across options
+- The agent **MUST** verify that risk/reward profiles span a meaningful range from conservative to aggressive
+- The agent **MUST** verify that each option has a credible theory of change, not just aspirational projections

package/studios/finance/STUDIO.md

@@ -0,0 +1,11 @@
+---
+name: finance
+description: Financial planning, budgeting, and analysis lifecycle
+stages: [forecast, budget, analysis, reporting, close]
+category: back-office
+default_model: sonnet
+---
+
+# Finance Studio
+
+Financial planning and analysis lifecycle covering forecasting, budgeting, variance analysis, reporting, and period close. Supports annual planning cycles, quarterly reviews, and ad-hoc financial analysis. Uses git persistence for auditability and version-controlled financial models.

package/studios/finance/examples/2026-q3-forecast/intent.md

@@ -0,0 +1,17 @@
+---
+title: 2026-q3-forecast
+studio: finance
+stages: [forecast, budget, analysis, reporting, close]
+status: complete
+---
+
+# 2026 Q3 Forecast
+
+Quarterly forecast for Q3 2026 — revenue, opex, and cash runway. Base case uses pipeline coverage (not raw pipeline). Known events: 5 senior engineering hires start in July; Acme renewal lands in August.
+
+## Goals
+
+- Revenue forecast using pipeline × historical coverage ratio (not raw pipeline sum)
+- Headcount-adjusted opex reflecting the 5-hire July ramp
+- Cash runway model with low/base/high scenarios
+- Variance walk from Q2 actuals to Q3 forecast for the board pack

package/studios/finance/examples/2026-q3-forecast/pre-intent.yml

@@ -0,0 +1,17 @@
+related:
+  - executive-strategy
+  - sales
+  - product-strategy
+turns:
+  - role: user
+    text: "Build the Q3 2026 forecast — revenue, opex, and cash runway."
+  - role: agent
+    text: "Base case on current pipeline only, or pipeline + pipeline coverage adjustment?"
+  - role: user
+    text: "Pipeline + coverage. Our win rate normalizes the top-of-funnel."
+  - role: agent
+    text: "Any known one-time items landing in Q3 — hires, renewals, big deals?"
+  - role: user
+    text: "5 senior eng hires starting in July, and the Acme renewal in August."
+  - role: agent
+    text: "Got it — Q3 model keyed off pipeline-coverage revenue, with the headcount and renewal events explicit."

package/studios/finance/fix-hats/reconciler.md

@@ -0,0 +1,10 @@
+---
+agent_type: general-purpose
+---
+**Focus:** Reconcile cross-stage artifacts against studio-wide standards. You are NOT wearing a stage-specific hat — you are resolving a whole-intent finding that spans stages. Your mandate is alignment, not fresh authoring.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** create new units, new scope, or new features
+- The agent **MUST NOT** modify unit workflow engine fields
+- The agent **MUST NOT** touch artifacts unrelated to the named finding
+- The agent **MUST NOT** re-open decisions settled at each stage's review gate

package/studios/finance/fix-hats/validator.md

@@ -0,0 +1,10 @@
+---
+agent_type: general-purpose
+model: haiku
+---
+**Focus:** Independently verify that the reconciler's fix resolves the intent-scope feedback. You are the terminal hat — the workflow engine trusts your closure decision.
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** edit any file — you verify, you do not fix
+- The agent **MUST NOT** close a finding that isn't actually resolved
+- The agent **MUST NOT** reject a finding because "it's not worth fixing" — either close, leave open, or reject as genuinely invalid

package/studios/finance/review-agents/cross-stage-consistency.md

@@ -0,0 +1,17 @@
+---
+scope: intent
+interpretation: lens
+---
+**Mandate:** Verify the intent's artifacts are internally consistent across stages. You are the only reviewer that sees the whole intent at once — your job is to catch seams that per-stage reviewers miss.
+
+**Check:**
+- The agent **MUST** verify that each stage's outputs align with what upstream stages specified — no dropped requirements, no silent scope expansion
+- The agent **MUST** verify that naming is consistent across stages — a concept named one thing upstream should carry the same name downstream
+- The agent **MUST** verify that stages' declared outputs exist at the paths their unit frontmatter promised
+- The agent **MUST** verify that the stages collectively deliver the intent's stated goal (read `intent.md`) — partial delivery is a finding
+- The agent **MUST** verify that concerns raised by any stage's review were actually addressed (not silently ignored)
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** re-litigate decisions already approved at each stage's gate
+- The agent **MUST NOT** propose new features or scope additions
+- The agent **MUST NOT** flag stylistic preferences — concrete divergence only

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

@@ -0,0 +1,17 @@
+---
+name: analysis
+description: Perform variance analysis and track financial performance
+hats: [analyst, auditor]
+fix_hats: [analyst, feedback-assessor]
+review: auto
+elaboration: autonomous
+inputs:
+  - stage: budget
+    discovery: budget-plan
+  - stage: forecast
+    discovery: forecast-model
+---
+
+# Analysis
+
+Perform variance analysis and track financial performance.

package/studios/finance/stages/analysis/discovery/VARIANCE-REPORT.md

@@ -0,0 +1,29 @@
+---
+name: variance-report
+location: .haiku/intents/{intent-slug}/knowledge/VARIANCE-REPORT.md
+scope: intent
+format: text
+required: true
+---
+
+# Variance Report
+
+Analysis of actual financial results against budget and forecast, with root cause analysis and corrective action recommendations.
+
+## Content Guide
+
+Structure the report around understanding and acting on deviations:
+
+- **Variance summary** -- material variances by department or cost center with magnitude and direction
+- **Root cause analysis** -- each material variance classified as structural, timing, or operational with evidence
+- **Performance trends** -- multi-period comparison showing trajectory
+- **Corrective actions** -- specific, actionable recommendations for each material deviation
+- **Leading indicators** -- early warning signals for potential future variances
+- **Methodology** -- how variances were calculated and materiality thresholds applied
+
+## Quality Signals
+
+- Every material variance has a documented root cause supported by evidence
+- Corrective actions are specific enough to implement, not generic recommendations
+- Both favorable and unfavorable variances are analyzed
+- Trend analysis uses enough periods to distinguish signal from noise

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

@@ -0,0 +1,13 @@
+**Focus:** Perform variance analysis comparing actual financial results against budget and forecast, identifying root causes and trends.
+
+**Responsibilities:**
+- Calculate variances at the appropriate granularity (department, cost center, line item)
+- Classify variances as structural, timing, or operational with supporting evidence
+- Identify performance trends using multi-period comparisons
+- Recommend corrective actions for material deviations
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** report variances without root cause analysis
+- The agent **MUST NOT** treat all variances as equally important regardless of materiality
+- The agent **MUST NOT** ignore favorable variances that may indicate budget padding or timing issues
+- The agent **MUST NOT** present numbers without narrative context for decision-makers

package/studios/finance/stages/analysis/hats/auditor.md

@@ -0,0 +1,13 @@
+**Focus:** Verify the accuracy of financial data, validate analytical methodology, and ensure findings are supported by evidence.
+
+**Responsibilities:**
+- Cross-check data sources to confirm numbers used in analysis are accurate
+- Validate variance calculation methodology is consistent across periods and departments
+- Verify that root cause attributions are supported by evidence, not assumption
+- Flag any accounting irregularities or data quality issues
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** accept analyst conclusions without independently verifying the underlying data
+- The agent **MUST NOT** apply inconsistent materiality thresholds across different areas
+- The agent **MUST NOT** focus only on numerical accuracy while ignoring methodological soundness
+- The agent **MUST NOT** rubber-stamp analysis without substantive review

package/studios/finance/stages/analysis/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/analysis/outputs/VARIANCE-REPORT.md

@@ -0,0 +1,25 @@
+---
+name: variance-report
+location: .haiku/intents/{intent-slug}/stages/analysis/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Variance Report
+
+Financial performance analysis with root cause explanations for budget deviations.
+
+## Expected Artifacts
+
+- **Variance analysis** -- every significant deviation explained with root cause and corrective action
+- **Performance metrics** -- leading and lagging indicators with trend analysis
+- **Finding categorization** -- each finding classified as structural, timing, or operational
+- **Corrective actions** -- recommended responses for each significant variance
+
+## Quality Signals
+
+- Every deviation greater than threshold is explained with root cause
+- Both leading and lagging indicators are tracked with multi-period trends
+- Findings are categorized to distinguish self-correcting from action-required items
+- Corrective actions are specific and assigned

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

@@ -0,0 +1,16 @@
+# Analysis Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Variance report explains every deviation greater than 5% from budget with root cause and corrective action"
+- "Performance metrics include both leading and lagging indicators with trend analysis over at least 3 periods"
+- "Each finding is categorized as structural (requires budget revision), timing (self-correcting), or operational (requires action)"
+
+### Bad criteria — vague (no clear check)
+
+- "Variances are explained"
+- "Performance is tracked"
+- "Analysis is complete"
+

package/studios/finance/stages/analysis/review-agents/accuracy.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify variance analysis is accurate and root causes are evidence-based.
+
+**Check:**
+- The agent **MUST** verify that variance calculations are mathematically correct and use consistent methodology
+- The agent **MUST** verify that root cause attributions are supported by evidence, not assumption
+- The agent **MUST** verify that materiality thresholds are applied consistently across departments
+- The agent **MUST** verify that corrective action recommendations are specific and actionable

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

@@ -0,0 +1,15 @@
+---
+name: budget
+description: Allocate resources and set financial targets
+hats: [budget-owner, allocator, verifier]
+fix_hats: [budget-owner, feedback-assessor]
+review: external
+elaboration: collaborative
+inputs:
+  - stage: forecast
+    discovery: forecast-model
+---
+
+# Budget
+
+Allocate resources and set financial targets.

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

@@ -0,0 +1,29 @@
+---
+name: budget-plan
+location: .haiku/intents/{intent-slug}/knowledge/BUDGET-PLAN.md
+scope: intent
+format: text
+required: true
+---
+
+# Budget Plan
+
+Resource allocation framework with detailed allocations by department or cost center, each traceable to forecast assumptions and strategic objectives.
+
+## Content Guide
+
+Structure the budget plan for allocation and tracking:
+
+- **Budget envelope** -- total available resources with source (forecast scenario used)
+- **Allocation by department/cost center** -- detailed allocations with rationale for each
+- **Strategic alignment** -- how each allocation maps to strategic objectives
+- **Targets** -- quantified performance targets with measurement criteria
+- **Contingency reserves** -- size, rationale, and release conditions
+- **Resource validation** -- confirmation of resource availability against commitments
+
+## Quality Signals
+
+- Allocations sum within the approved envelope with variance explanations
+- Every line item traces to a forecast assumption or strategic objective
+- Contingency reserves are sized based on historical variance or risk analysis
+- Targets are quantified with clear measurement criteria