haiku-method 2.2.0

package/studios/documentation/stages/review/review-agents/completeness.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify the documentation fully addresses the gaps identified in the audit.
+
+**Check:**
+- The agent **MUST** verify that every priority gap from the audit stage has corresponding documentation
+- The agent **MUST** verify that no placeholder text, TODO markers, or incomplete sections remain
+- The agent **MUST** verify that all promised cross-references and links resolve correctly
+- The agent **MUST** verify that the documentation is self-contained — a reader does not need tribal knowledge to follow it

package/studios/documentation/templates/api-documentation.md

@@ -0,0 +1,38 @@
+---
+name: api-documentation
+studio: documentation
+description: Document a new or updated API
+parameters:
+  - name: api
+    description: API name or module being documented
+    required: true
+  - name: openapi_url
+    description: Link to OpenAPI spec if available
+    required: false
+units:
+  - name: "coverage-audit"
+    stage: audit
+    criteria:
+      - "All {{ api }} endpoints inventoried"
+      - "Existing documentation gaps identified"
+      - "Priority ranked by endpoint usage"
+  - name: "structure"
+    stage: outline
+    criteria:
+      - "Information architecture follows task-based organization"
+      - "Each endpoint has: description, parameters, examples, errors"
+  - name: "write"
+    stage: draft
+    criteria:
+      - "All endpoints documented with working code examples"
+      - "Authentication and error handling sections complete"
+      - "Examples tested against current API version"
+  - name: "review"
+    stage: review
+    criteria:
+      - "Technical accuracy verified against implementation"
+      - "Code examples compile and produce expected output"
+      - "No jargon without definition"
+---
+
+API documentation from audit through publish.

package/studios/executive-strategy/STUDIO.md

@@ -0,0 +1,11 @@
+---
+name: executive-strategy
+description: Executive decision-making and strategic planning lifecycle
+stages: [landscape, options, evaluate, decide, communicate]
+category: back-office
+default_model: sonnet
+---
+
+# Executive Strategy Studio
+
+Executive decision-making lifecycle for strategic initiatives. Covers market landscape analysis, option generation, tradeoff evaluation, structured decision-making, and stakeholder communication. Works for market entry decisions, organizational restructuring, investment decisions, and strategic pivots.

package/studios/executive-strategy/examples/ai-platform-strategy/intent.md

@@ -0,0 +1,17 @@
+---
+title: ai-platform-strategy
+studio: executive-strategy
+stages: [landscape, options, evaluate, decide, communicate]
+status: complete
+---
+
+# AI Platform vs. Portfolio Strategy
+
+Frame and decide whether the company pursues a unified AI platform strategy or doubles down on the two revenue-concentrated flagship products. Decision memo due at the next board meeting (6 weeks out). Current revenue is 70% concentrated in two flagships; the remaining 30% is spread across six smaller products.
+
+## Goals
+
+- Landscape scan of platform vs. portfolio plays by direct and adjacent competitors
+- Three concrete options with revenue, cost, and timeline modeling
+- Clear recommendation tied to the 70/30 revenue concentration reality
+- Board-ready communication deck and pre-read memo

package/studios/executive-strategy/examples/ai-platform-strategy/pre-intent.yml

@@ -0,0 +1,17 @@
+related:
+  - product-strategy
+  - marketing
+  - finance
+turns:
+  - role: user
+    text: "Board wants a platform-vs-point-solution answer on our AI roadmap."
+  - role: agent
+    text: "What's the current revenue split between the point products?"
+  - role: user
+    text: "70% on two flagship products, 30% on six smaller ones."
+  - role: agent
+    text: "Timeline — board decision this quarter, or longer?"
+  - role: user
+    text: "Next board meeting. 6 weeks out."
+  - role: agent
+    text: "Got it — I'll frame platform vs. portfolio tradeoffs against that revenue concentration."

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

@@ -0,0 +1,15 @@
+---
+name: communicate
+description: Develop stakeholder communications and rollout plan
+hats: [communicator, planner, verifier]
+fix_hats: [communicator, feedback-assessor]
+review: ask
+elaboration: autonomous
+inputs:
+  - stage: decide
+    discovery: decision-brief
+---
+
+# Communicate
+
+Develop stakeholder communications and rollout plan.

package/studios/executive-strategy/stages/communicate/hats/communicator.md

@@ -0,0 +1,13 @@
+**Focus:** Craft stakeholder communications that convey the strategic decision clearly, consistently, and with appropriate context for each audience.
+
+**Responsibilities:**
+- Develop messaging tailored to each stakeholder group's concerns and information needs
+- Ensure messaging consistency across all audiences and channels
+- Anticipate questions and objections with pre-prepared responses
+- Create communication materials (presentations, memos, FAQs) for each channel
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** use identical messaging for all audiences regardless of their concerns
+- The agent **MUST NOT** communicat the decision without explaining the rationale
+- The agent **MUST NOT** b evasive about risks or tradeoffs, which erodes trust when they surface later
+- The agent **MUST NOT** treat communication as a one-way broadcast rather than a dialogue

package/studios/executive-strategy/stages/communicate/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/communicate/hats/planner.md

@@ -0,0 +1,13 @@
+**Focus:** Design the rollout plan that translates the strategic decision into a sequenced set of actions with ownership and milestones.
+
+**Responsibilities:**
+- Sequence rollout actions with dependencies and ownership assignments
+- Define measurable milestones that track progress toward strategic objectives
+- Identify resource requirements and confirm availability for each action
+- Build contingency plans for the highest-risk elements of the rollout
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** create a plan that is too high-level to be actionable
+- The agent **MUST NOT** assign actions without confirming the owner has capacity and authority
+- The agent **MUST** identify dependencies between rollout actions
+- The agent **MUST NOT** plan the happy path without contingencies for likely obstacles

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

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit design/synthesis artifact for the communicate stage of executive-strategy. Units here are communication artifact — 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/communicate/outputs/COMMS-PACKAGE.md

@@ -0,0 +1,26 @@
+---
+name: comms-package
+location: .haiku/intents/{intent-slug}/stages/communicate/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Communications Package
+
+Stakeholder communications and rollout plan for implementing the strategic decision.
+
+## Content Guide
+
+- **Messaging framework** -- core messages adapted for each stakeholder group
+- **Communication materials** -- presentations, memos, and FAQs for each channel
+- **Rollout plan** -- sequenced actions with owners, dependencies, and milestones
+- **FAQ document** -- anticipated questions with pre-approved responses
+- **Communication calendar** -- timing and channel for each stakeholder touchpoint
+
+## Quality Signals
+
+- Core messaging is consistent across all audiences and channels
+- Each audience receives context appropriate to their concerns
+- Rollout plan has specific owners and measurable milestones
+- FAQ covers the most likely questions from each stakeholder group

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

@@ -0,0 +1,16 @@
+# Communicate Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Communication plan tailors messaging to each stakeholder group with specific channels, timing, and key messages"
+- "Rollout plan sequences actions with dependencies, owners, and measurable milestones"
+- "FAQ document anticipates the top 10 likely questions with pre-approved responses"
+
+### Bad criteria — vague (no clear check)
+
+- "Communication is planned"
+- "Stakeholders are informed"
+- "Rollout is ready"
+

package/studios/executive-strategy/stages/communicate/review-agents/consistency.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify communications are consistent across audiences and the rollout plan is actionable.
+
+**Check:**
+- The agent **MUST** verify that core messaging is consistent across all stakeholder communications
+- The agent **MUST** verify that each audience receives context appropriate to their concerns and decision role
+- The agent **MUST** verify that rollout plan has specific owners, dependencies, and measurable milestones
+- The agent **MUST** verify that fAQ anticipates likely questions and provides pre-approved responses

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

@@ -0,0 +1,17 @@
+---
+name: decide
+description: Apply decision framework and produce recommendation
+hats: [advisor, facilitator, verifier]
+fix_hats: [advisor, feedback-assessor]
+review: external
+elaboration: collaborative
+inputs:
+  - stage: evaluate
+    discovery: evaluation-report
+  - stage: options
+    discovery: options-matrix
+---
+
+# Decide
+
+Apply decision framework and produce recommendation.

package/studios/executive-strategy/stages/decide/discovery/DECISION-BRIEF.md

@@ -0,0 +1,29 @@
+---
+name: decision-brief
+location: .haiku/intents/{intent-slug}/knowledge/DECISION-BRIEF.md
+scope: intent
+format: text
+required: true
+---
+
+# Decision Brief
+
+Recommendation with supporting rationale, risk acknowledgment, and decision record.
+
+## Content Guide
+
+Structure the brief for clear, defensible decision-making:
+
+- **Recommendation** -- the recommended option with concise rationale
+- **Supporting arguments** -- the strongest evidence and reasoning for the recommendation
+- **Counterarguments** -- the strongest arguments against, addressed directly
+- **Risk acknowledgment** -- known risks of the recommended option with mitigation plans
+- **Dissenting perspectives** -- stakeholder disagreements documented with their reasoning
+- **Next steps** -- what implementing the recommendation requires
+
+## Quality Signals
+
+- Recommendation follows logically from the evaluation evidence
+- Counterarguments are addressed directly, not minimized or dismissed
+- Dissenting perspectives are documented respectfully with their reasoning
+- Next steps are specific enough to begin implementation

package/studios/executive-strategy/stages/decide/hats/advisor.md

@@ -0,0 +1,13 @@
+**Focus:** Synthesize evaluation results into a clear recommendation with supporting rationale and acknowledged risks.
+
+**Responsibilities:**
+- Formulate a recommendation that follows logically from the evaluation evidence
+- Present the strongest arguments for and against the recommended option
+- Acknowledge risks and limitations rather than minimizing them
+- Ensure the recommendation is actionable with clear next steps
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** make recommendations based on preference rather than evaluation evidence
+- The agent **MUST NOT** present only the case for the recommendation without counterarguments
+- The agent **MUST NOT** bury risks and limitations in appendices rather than addressing them directly
+- The agent **MUST NOT** recommend without specifying what "implementing this recommendation" means

package/studios/executive-strategy/stages/decide/hats/facilitator.md

@@ -0,0 +1,13 @@
+**Focus:** Ensure the decision process is inclusive, transparent, and produces a decision that stakeholders can commit to.
+
+**Responsibilities:**
+- Ensure all relevant stakeholder perspectives are heard and considered
+- Document areas of agreement and disagreement transparently
+- Facilitate conflict resolution when stakeholders disagree on the recommendation
+- Capture dissenting views with their rationale for the decision record
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** rush to consensus without genuinely exploring disagreements
+- The agent **MUST NOT** allow the most senior voice to override evidence-based analysis
+- The agent **MUST NOT** document only the final decision without the deliberation that led to it
+- The agent **MUST NOT** exclud stakeholders whose perspectives are uncomfortable but relevant

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

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit operational artifact for the decide stage of executive-strategy. Units here are decision ratification — 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/executive-strategy/stages/decide/outputs/DECISION-BRIEF.md

@@ -0,0 +1,25 @@
+---
+name: decision-brief
+location: .haiku/intents/{intent-slug}/stages/decide/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Decision Brief
+
+Clear recommendation with supporting arguments, counterarguments, and decision framework documentation.
+
+## Expected Artifacts
+
+- **Recommendation** -- clear recommendation with the strongest supporting and counterarguments
+- **Decision framework** -- criteria weights, who set them, and how conflicts were resolved
+- **Dissenting perspectives** -- documented with their reasoning, not dismissed
+- **Decision record** -- what was decided, by whom, and the rationale
+
+## Quality Signals
+
+- Recommendation presents both supporting arguments and counterarguments
+- Decision framework documents criteria weights and who set them
+- Dissenting perspectives are documented with reasoning
+- Decision is recorded with clear rationale and accountability

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

@@ -0,0 +1,16 @@
+# Decide Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Decision brief presents a clear recommendation with the 3 strongest supporting arguments and the 2 strongest counterarguments"
+- "Decision framework documents the criteria weights, who set them, and how conflicts were resolved"
+- "Dissenting perspectives are documented with their reasoning, not dismissed or omitted"
+
+### Bad criteria — vague (no clear check)
+
+- "Decision is made"
+- "Recommendation is clear"
+- "Stakeholders agree"
+

package/studios/executive-strategy/stages/decide/review-agents/transparency.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify the decision process is transparent and the recommendation is supported by evidence.
+
+**Check:**
+- The agent **MUST** verify that recommendation follows logically from the evaluation evidence
+- The agent **MUST** verify that counterarguments are addressed directly, not dismissed
+- The agent **MUST** verify that dissenting perspectives are documented with their reasoning
+- The agent **MUST** verify that decision criteria and their application are transparent and auditable

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

@@ -0,0 +1,17 @@
+---
+name: evaluate
+description: Analyze tradeoffs and model scenarios for each option
+hats: [evaluator, risk-analyst, verifier]
+fix_hats: [evaluator, feedback-assessor]
+review: ask
+elaboration: collaborative
+inputs:
+  - stage: options
+    discovery: options-matrix
+  - stage: landscape
+    discovery: landscape-analysis
+---
+
+# Evaluate
+
+Analyze tradeoffs and model scenarios for each option.

package/studios/executive-strategy/stages/evaluate/discovery/EVALUATION-REPORT.md

@@ -0,0 +1,29 @@
+---
+name: evaluation-report
+location: .haiku/intents/{intent-slug}/knowledge/EVALUATION-REPORT.md
+scope: intent
+format: text
+required: true
+---
+
+# Evaluation Report
+
+Multi-criteria scoring, scenario analysis, and risk assessment for each strategic option.
+
+## Content Guide
+
+Structure the report to support decision-making:
+
+- **Evaluation framework** -- criteria, weights, and scoring methodology
+- **Scoring results** -- each option scored against all criteria with documented reasoning
+- **Scenario analysis** -- each option modeled under bull, base, and bear conditions
+- **Risk assessment** -- top risks per option with probability, impact, and mitigation
+- **Tradeoff analysis** -- key tradeoffs between options that scoring alone cannot resolve
+- **Comparative summary** -- high-level comparison enabling informed decision-making
+
+## Quality Signals
+
+- Evaluation criteria and weights reflect stakeholder-validated priorities
+- Scoring rationale is documented for each criterion, not just the composite score
+- Scenario modeling covers genuinely adverse conditions
+- Tradeoffs are identified explicitly where they exist

package/studios/executive-strategy/stages/evaluate/hats/evaluator.md

@@ -0,0 +1,13 @@
+**Focus:** Apply multi-criteria evaluation to score and compare strategic options using a consistent, transparent framework.
+
+**Responsibilities:**
+- Define evaluation criteria with explicit weights reflecting strategic priorities
+- Score each option against all criteria with documented reasoning
+- Identify tradeoffs between options that cannot be resolved by scoring alone
+- Produce a comparative summary that enables informed decision-making
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** weight criteria after seeing the scores to justify a preferred option
+- The agent **MUST NOT** treat all criteria as equally important without stakeholder input on priorities
+- The agent **MUST NOT** reduc complex tradeoffs to a single composite score that hides important nuances
+- The agent **MUST** document the reasoning behind individual scores

package/studios/executive-strategy/stages/evaluate/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/evaluate/hats/risk-analyst.md

@@ -0,0 +1,13 @@
+**Focus:** Assess risks for each strategic option, stress-test key assumptions, and model downside scenarios.
+
+**Responsibilities:**
+- Identify the top risks for each option with probability and impact estimates
+- Stress-test key financial and operational assumptions under adverse conditions
+- Model downside scenarios to quantify worst-case outcomes
+- Recommend risk mitigation strategies and their cost/feasibility
+
+**Anti-patterns (RFC 2119):**
+- The agent **MUST NOT** list risks without quantifying their probability or impact
+- The agent **MUST NOT** stress-test only the most obvious assumptions while ignoring hidden dependencies
+- The agent **MUST NOT** present risk analysis that makes all options look equally risky
+- The agent **MUST** connect risk mitigation recommendations to specific, actionable plans

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

@@ -0,0 +1,26 @@
+**Focus:** Validate the per-unit knowledge artifact for the evaluate stage of executive-strategy. Units here are option evaluation — 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/evaluate/outputs/EVALUATION-REPORT.md

@@ -0,0 +1,25 @@
+---
+name: evaluation-report
+location: .haiku/intents/{intent-slug}/stages/evaluate/artifacts/
+scope: intent
+format: text
+required: true
+---
+
+# Evaluation Report
+
+Tradeoff analysis and scenario modeling results for each strategic option.
+
+## Expected Artifacts
+
+- **Tradeoff analysis** -- each option scored against weighted criteria with explicit reasoning
+- **Scenario modeling** -- options tested under multiple market conditions with quantified outcomes
+- **Risk analysis** -- top risks per option with probability estimates and mitigation strategies
+- **Comparative summary** -- side-by-side comparison highlighting key differentiators
+
+## Quality Signals
+
+- Each option is scored against weighted criteria with reasoning, not just numbers
+- Scenarios cover at least 3 market conditions (bull, base, bear)
+- Top 3 risks per option have probability estimates and mitigations
+- Analysis is objective and does not pre-select a winner

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

@@ -0,0 +1,16 @@
+# Evaluate Stage — Elaboration
+
+## Criteria Guidance
+
+### Good criteria — concrete and verifiable
+
+- "Tradeoff analysis scores each option against weighted criteria with explicit reasoning for each score"
+- "Scenario modeling tests each option under at least 3 market conditions (bull, base, bear) with quantified outcomes"
+- "Risk analysis identifies the top 3 risks per option with probability estimates and mitigation strategies"
+
+### Bad criteria — vague (no clear check)
+
+- "Options are evaluated"
+- "Tradeoffs are analyzed"
+- "Risks are identified"
+

package/studios/executive-strategy/stages/evaluate/review-agents/objectivity.md

@@ -0,0 +1,10 @@
+---
+interpretation: lens
+---
+**Mandate:** The agent **MUST** verify evaluation is objective and not biased toward a pre-determined outcome.
+
+**Check:**
+- The agent **MUST** verify that evaluation criteria and weights were defined before scoring, not adjusted afterward
+- The agent **MUST** verify that scoring rationale is documented with evidence for each criterion
+- The agent **MUST** verify that scenario modeling covers genuinely adverse conditions, not just minor variations
+- The agent **MUST** verify that risk analysis is honest about downside exposure, not minimized