convoke-agents 2.0.0

package/_bmad/bme/_vortex/workflows/learning-card/steps/step-05-implications.md

@@ -0,0 +1,134 @@
+---
+step: 5
+workflow: learning-card
+title: Implications
+---
+
+# Step 5: Implications
+
+Now we translate validated learnings into strategic implications. What does this mean for the product? What decisions does this inform? This is where learning becomes action.
+
+## Why This Matters
+
+A learning card without implications is like a lab report without a conclusion section. Implications:
+- Connect experiments to product strategy
+- Inform resource allocation and prioritization
+- Trigger pivot/patch/persevere decisions
+- Prevent the "interesting but so what?" problem
+
+## Your Task
+
+### 1. Product Implications
+
+For each validated learning, answer: what does this mean for what we build?
+
+**Framework for product implications:**
+- **Build:** This learning tells us we SHOULD build X because Y
+- **Don't build:** This learning tells us we should NOT build X because Y
+- **Modify:** This learning tells us we should CHANGE our approach to X
+- **Investigate further:** We can't decide yet; we need more evidence on X
+
+Be specific. "We should improve the product" is not an implication. "We should add a team champion onboarding flow because adoption requires a daily role model" is.
+
+### 2. Strategy Implications
+
+Zoom out from the product to the business:
+
+- **Value proposition:** Does this learning change how we describe our value?
+- **Target market:** Does this learning narrow, expand, or shift our target user?
+- **Business model:** Does this affect pricing, distribution, or revenue assumptions?
+- **Competitive positioning:** Does this change how we differentiate?
+
+### 3. Assumption Cascade
+
+Every learning changes the assumption landscape. Map the ripple effects:
+
+**Assumptions now STRONGER (more confident):**
+- List assumptions that this learning reinforces
+- Note why and how much more confident you are
+
+**Assumptions now WEAKER (less confident):**
+- List assumptions that this learning undermines
+- Note what needs to happen to resolve them
+
+**NEW assumptions to test:**
+- What new assumptions emerged from this learning?
+- How important are they to your strategy?
+- What's the fastest way to test them?
+
+### 4. Decision Trigger Assessment
+
+Does this learning trigger a major decision?
+
+**Use this decision readiness framework:**
+
+| Signal | Meaning | Action |
+|--------|---------|--------|
+| Multiple learnings point the same direction | Pattern emerging | Consider acting on it |
+| Key assumption invalidated | Strategy at risk | Trigger pivot/patch/persevere review |
+| Core hypothesis validated | Green light | Proceed to next stage with confidence |
+| Evidence is mixed | Insufficient clarity | Design follow-up experiment |
+| Evidence contradicts team consensus | Conflict between belief and data | Facilitate evidence-based discussion |
+
+### 5. Recommended Next Actions
+
+Based on the implications, what should happen next? Be concrete:
+
+- **Immediately:** What should change RIGHT NOW based on this learning?
+- **This sprint:** What experiments or work items should be added?
+- **This quarter:** What strategic adjustments should be considered?
+- **Flag for review:** What should be raised with stakeholders?
+
+## Example
+
+**Product Implications:**
+
+1. **BUILD:** Add a "team champion" onboarding path that designates one person per team as the daily role model. Learning shows champions drive 25%+ higher adoption.
+
+2. **DON'T BUILD:** Don't invest in reminder/notification features as the primary adoption mechanism. Team C had reminders but still dropped to 62% without a champion. Habit formation is social, not mechanical.
+
+3. **MODIFY:** Target cross-timezone teams first, not co-located teams. The value proposition is significantly stronger when synchronous meetings are painful (timezone friction) vs. merely inefficient (co-located).
+
+4. **INVESTIGATE:** We need to understand whether 2-week novelty drop-off stabilizes or continues declining before committing to a full MVP build.
+
+**Strategy Implications:**
+
+- **Value proposition shift:** From "replace standups" to "enable effective async collaboration for distributed teams." The timezone angle is stronger than the meeting-hate angle.
+- **Target market narrowed:** Focus on teams with 3+ time zones represented, not all remote teams. This is a smaller market but much higher-urgency problem.
+- **Business model unchanged:** No impact on pricing assumptions yet -- need to test willingness to pay separately.
+
+**Assumption Cascade:**
+
+Stronger:
+- "Distributed teams need async status tools" (HIGH confidence now)
+- "Video is more information-rich than text status" (MEDIUM-HIGH)
+
+Weaker:
+- "All remote teams are our market" (now believe only cross-timezone teams are initial market)
+- "Consistent daily updates will happen naturally" (requires social mechanism)
+
+New:
+- "Team champion role can be onboarded, not just organic" (need to test with non-volunteer champions)
+- "4+ weeks needed for habit formation" (need longer experiment)
+
+**Decision Trigger:** This learning triggers a SCOPE NARROWING decision, not a pivot. The core hypothesis is partially validated but the target market needs to be narrowed.
+
+**Recommended Next Actions:**
+- **Immediately:** Update our target persona to specify "cross-timezone teams"
+- **This sprint:** Design a 4-week replication study with 8-10 teams, including assigned champions
+- **This quarter:** Revisit co-located teams after nailing the distributed use case
+- **Flag for review:** Share the market narrowing with founding team -- smaller initial TAM but higher conversion potential
+
+---
+
+## Your Turn
+
+Please map the implications of your validated learnings using the structure above.
+
+**Key principle:** Implications should be specific enough that someone could act on them without asking you for clarification. If your implication is "we should think about this more," push yourself to say WHAT you should think about and HOW.
+
+## Next Step
+
+When you've mapped the implications, I'll load:
+
+{project-root}/_bmad/bme/_vortex/workflows/learning-card/steps/step-06-synthesize.md

package/_bmad/bme/_vortex/workflows/learning-card/steps/step-06-synthesize.md

@@ -0,0 +1,121 @@
+---
+step: 6
+workflow: learning-card
+title: Synthesize
+---
+
+# Step 6: Synthesize
+
+Let's bring everything together into the final learning card artifact -- a permanent record of what was tested, what was learned, and what it means.
+
+## Why This Matters
+
+The synthesized learning card serves as:
+- An institutional knowledge asset that outlives team changes
+- Input for pivot/patch/persevere decisions
+- Evidence for stakeholder discussions
+- A building block for compounding organizational learning
+- A reference point for future experiments on the same topic
+
+## Your Task
+
+Before I generate the final artifact, I need a few synthesis elements:
+
+### 1. One-Sentence Learning Summary
+
+Distill the entire learning card into ONE sentence a busy executive could read:
+
+**Format:** "We [validated/invalidated/partially validated] that [core hypothesis], discovering that [key insight], which means [primary implication]."
+
+**Examples:**
+- "We partially validated that distributed teams will adopt async video updates, discovering that adoption requires a team champion role model, which means our onboarding must include champion designation before team rollout."
+- "We invalidated that price sensitivity is our biggest conversion barrier, discovering that trust and perceived reliability matter more, which means our landing page should lead with social proof, not discounts."
+
+### 2. Confidence Rating
+
+Rate your overall confidence in the learnings:
+
+| Rating | Meaning |
+|--------|---------|
+| **HIGH** | Strong evidence, adequate sample, controlled experiment, replicable |
+| **MEDIUM** | Directional evidence, some limitations, patterns are clear but not proven |
+| **LOW** | Suggestive evidence, significant limitations, needs replication |
+| **EXPLORATORY** | Early signal only, hypothesis-generating not hypothesis-confirming |
+
+### 3. Links to Other Learning Cards
+
+Does this learning connect to other experiments or learning cards?
+
+- **Builds on:** Which previous learnings does this extend?
+- **Contradicts:** Does this conflict with any previous findings?
+- **Enables:** What future experiments does this make possible?
+
+### 4. Tags for Discoverability
+
+Tag the learning card so future team members can find it:
+
+Suggested tag categories:
+- **Topic:** (e.g., onboarding, pricing, engagement, retention, acquisition)
+- **User segment:** (e.g., remote-managers, enterprise, SMB)
+- **Vortex stream:** (e.g., Empathize, Externalize, Systematize)
+- **Experiment type:** (e.g., survey, A/B-test, interview, prototype)
+- **Outcome:** (e.g., validated, invalidated, inconclusive)
+
+## Example
+
+**One-Sentence Summary:** "We partially validated that distributed teams will adopt async video updates, discovering that a team champion role model drives 25%+ higher adoption than reminders alone, which means our MVP must include champion onboarding as a launch prerequisite."
+
+**Confidence:** MEDIUM - Clear directional patterns from 3 teams/19 participants over 2 weeks, but sample is too small and duration too short for high confidence. Replication planned.
+
+**Links:**
+- **Builds on:** Learning Card: "Remote Manager Pain Points" (validated that status tracking is top time-waster)
+- **Contradicts:** None
+- **Enables:** Follow-up experiments on champion onboarding mechanics, long-term habit retention, pricing for distributed teams
+
+**Tags:** `async-collaboration`, `distributed-teams`, `video-updates`, `adoption`, `Externalize`, `concierge-MVP`, `partially-validated`
+
+---
+
+## Your Turn
+
+Please provide:
+1. Your one-sentence learning summary
+2. Your confidence rating with justification
+3. Any links to related learning cards
+4. Tags for discoverability
+
+## Final Step
+
+When you've provided these synthesis elements, I'll:
+1. Generate your complete learning card artifact using the template
+2. Save it to `{output_folder}/learning-card-{experiment-name}-{date}.md`
+3. Highlight any follow-up experiments you should consider
+4. Suggest whether this learning is ready to feed into a pivot/patch/persevere decision
+
+---
+
+## Workflow Complete
+
+After synthesis, your learning card artifact will include:
+- Complete experiment context and methodology
+- Raw results with data quality assessment
+- Rigorous analysis with alternative explanations considered
+- Validated learnings with confidence ratings
+- Strategic implications and recommended actions
+- Links to related learnings and discoverability tags
+- Revision history for future updates as you learn more
+
+## Vortex Compass
+
+Based on what you just completed, here are your evidence-driven options:
+
+| If you learned... | Consider next... | Agent | Why |
+|---|---|---|---|
+| Learning triggers strategic decision | pivot-patch-persevere | Max 🧭 | Decide: pivot, patch, or persevere |
+| Need more data | lean-experiment | Wade 🧪 | Run follow-up experiments |
+| Assumptions invalidated | contextualize-scope | Emma 🎯 | Re-frame the problem based on evidence (HC8) |
+
+> **Note:** These are evidence-based recommendations. You can navigate to any Vortex agent
+> at any time based on your judgment.
+
+**Or run Max's [VN] Vortex Navigation** for a full gap analysis across all streams.

package/_bmad/bme/_vortex/workflows/learning-card/validate.md

@@ -0,0 +1,134 @@
+---
+workflow: validate-learning-card
+type: single-file
+description: Validate an existing learning card for evidence quality, learning clarity, and implication actionability
+author: Max (learning-decision-expert)
+---
+
+# Validate Existing Learning Card
+
+Bring me a learning card and I'll help you assess whether the evidence is rigorous, the learnings are clear, and the implications are actionable.
+
+## Why Validation Matters
+
+Learning cards are only as valuable as the rigor behind them. A learning card that claims "validated" when the evidence is weak leads to bad decisions. A learning card that buries a critical insight in vague language wastes organizational knowledge. Validation ensures your learning cards are trustworthy decision-making inputs.
+
+## Validation Process
+
+### 1. Evidence Quality Check
+
+For each learning statement in the card, assess the evidence:
+
+**Strong Evidence:**
+- Pre-defined success criteria existed before the experiment
+- Sample size is adequate for the method (5+ interviews per segment, 30+ survey responses, statistical significance for A/B tests)
+- Confounding factors are identified and accounted for
+- Data quality limitations are acknowledged
+- Both quantitative and qualitative data align
+
+**Red Flags:**
+- Success criteria were defined AFTER seeing results (moving goalposts)
+- Sample size is too small for the claims being made
+- Only positive results are reported (cherry-picking)
+- No alternative explanations considered
+- Quantitative and qualitative data contradict each other without explanation
+- "We believe" instead of "The data shows"
+- No data quality assessment
+
+### 2. Learning Clarity Check
+
+For each validated learning, verify:
+
+**Is it specific?**
+- "Users prefer async" (too vague)
+- "Cross-timezone teams of 5-8 achieve 90%+ adoption of async video updates when a team champion records daily" (specific and actionable)
+
+**Is it bounded?**
+- Does it state the conditions under which the learning holds?
+- Does it acknowledge where it might NOT hold?
+- Is the confidence rating appropriate given the evidence?
+
+**Is it honest about uncertainty?**
+- Does it distinguish between "validated" and "partially validated"?
+- Are inconclusive results marked as such?
+- Are new questions raised by the findings documented?
+
+**Is it falsifiable?**
+- Could a future experiment prove this learning wrong?
+- If not, it's an opinion, not a learning
+
+### 3. Implication Actionability Check
+
+For each implication listed in the card:
+
+**Can someone act on this?**
+- "We should improve our onboarding" (not actionable)
+- "We should add a champion designation step to team onboarding within the first 48 hours" (actionable)
+
+**Is the logic sound?**
+- Does the implication logically follow from the learning?
+- Are there leaps in reasoning (learning says X, implication assumes Y)?
+
+**Are the next actions concrete?**
+- Do they have clear owners?
+- Do they have timeframes?
+- Do they have success criteria of their own?
+
+### 4. Completeness Check
+
+**Required Sections:**
+- [ ] Experiment context with hypothesis, method, and success criteria
+- [ ] Raw results (both quantitative and qualitative)
+- [ ] Analysis with pattern identification and alternative explanations
+- [ ] Validated learnings with confidence ratings
+- [ ] Implications with concrete next actions
+- [ ] Evidence quality self-assessment
+- [ ] Connection to other learning cards or experiments
+
+**Missing Perspectives:**
+- Are dissenting data points captured?
+- Are stakeholder concerns acknowledged?
+- Is the learning card useful to someone who WASN'T part of the experiment?
+
+---
+
+## Your Turn
+
+**Please share:**
+1. The learning card you want to validate (paste content or provide file path)
+2. Optionally: the raw experiment data for cross-reference
+
+I'll review it and provide:
+- **Evidence assessment** - Is the evidence strong enough to support the claims?
+- **Learning clarity score** - Are the learnings specific, bounded, and falsifiable?
+- **Implication readiness** - Are the implications actionable and logically grounded?
+- **Overall validation rating** - Ready for decisions / Needs revision / Insufficient evidence
+- **Specific improvement suggestions** - Exactly what to fix and how
+
+---
+
+## Validation Rating Scale
+
+**Ready for Decisions:**
+- Evidence is strong and well-documented
+- Learnings are specific, bounded, and appropriately confident
+- Implications are actionable with clear next steps
+- Limitations are honestly acknowledged
+- This learning card can safely inform strategic decisions
+
+**Needs Revision:**
+- Evidence has gaps that should be addressed
+- Some learnings are vague or overconfident
+- Implications don't clearly follow from learnings
+- Missing sections or incomplete analysis
+- Fix the identified issues before using for decisions
+
+**Insufficient Evidence:**
+- Claims exceed what the evidence supports
+- Critical methodology flaws (too small sample, no success criteria, severe bias)
+- Learnings are opinions dressed as validated findings
+- Requires additional experimentation before conclusions can be drawn
+
+---
+
+Ready to validate your learning card? Share it with me and I'll provide a rigorous, honest assessment.

package/_bmad/bme/_vortex/workflows/learning-card/workflow.md

@@ -0,0 +1,51 @@
+---
+workflow: learning-card
+type: step-file
+description: Capture validated learnings from experiments as structured learning cards
+author: Max (learning-decision-expert)
+version: 1.5.0
+---
+
+# Create Learning Card Workflow
+
+This workflow guides you through creating a learning card -- a structured record of what was tested, what was learned, and what it means for your product.
+
+## What is a Learning Card?
+
+A learning card is a structured lab report for Lean Startup experiments. It captures the full lifecycle of a learning: from the experiment that generated it, through the raw data, to the validated insight and its strategic implications. Learning cards:
+
+- **Preserve institutional knowledge** - Learnings survive team changes, pivots, and time
+- **Force rigorous analysis** - Separating raw data from interpretation prevents confirmation bias
+- **Create a decision trail** - Future you can trace WHY decisions were made
+- **Enable compounding learning** - Each card builds on previous ones, accelerating discovery
+- **Feed decision frameworks** - Learning cards are the input for pivot/patch/persevere decisions
+
+## Workflow Structure
+
+**Step-file architecture:**
+- Just-in-time loading (each step loads only when needed)
+- Sequential enforcement (must complete step N before step N+1)
+- State tracking in frontmatter (progress preserved)
+
+## Steps Overview
+
+1. **Experiment Context** - What experiment was run, why, and what hypothesis was tested?
+2. **Raw Results** - Capture raw data and observations without interpretation
+3. **Analysis** - Analyze what the data actually says: patterns, significance, surprises
+4. **Validated Learning** - Extract validated learnings: what assumptions were confirmed or invalidated?
+5. **Implications** - What does this mean for the product and what decisions does it inform?
+6. **Synthesize** - Create the final learning card artifact
+
+## Output
+
+**Artifact:** Learning card markdown file in `{output_folder}/learning-card-{experiment-name}-{date}.md`
+
+**Template:** Uses [learning-card.template.md](learning-card.template.md)
+
+---
+
+## INITIALIZATION
+
+Load config from {project-root}/_bmad/bme/_vortex/config.yaml
+
+Load step: {project-root}/_bmad/bme/_vortex/workflows/learning-card/steps/step-01-experiment-context.md

package/_bmad/bme/_vortex/workflows/mvp/mvp.template.md

@@ -0,0 +1,40 @@
+---
+title: "MVP Specification: {mvp-name}"
+date: {date}
+created-by: {user-name} with Wade (lean-experiments-specialist)
+type: mvp-spec
+status: HYPOTHESIS
+version: 1.0
+---
+
+# MVP Specification: {mvp-name}
+
+## Riskiest Assumption
+
+{riskiest-assumption}
+
+## Success Criteria
+
+{success-criteria}
+
+## MVP Scope
+
+### What's In
+{scope-in}
+
+### What's Out
+{scope-out}
+
+## Build-Measure-Learn Plan
+
+{bml-plan}
+
+## Timeline
+
+{timeline}
+
+---
+
+**Created with:** Convoke v2.0.0 - Vortex Pattern (Externalize Stream)
+**Agent:** Wade (Lean Experiments Specialist)
+**Workflow:** mvp

package/_bmad/bme/_vortex/workflows/mvp/steps/step-01-riskiest-assumption.md

@@ -0,0 +1,17 @@
+---
+step: 1
+workflow: mvp
+title: Identify Riskiest Assumption
+---
+
+# Step 1: Identify Riskiest Assumption
+
+What assumption, if wrong, would kill your idea?
+
+## Your Task
+
+Identify your riskiest assumption and why it's critical.
+
+## Next Step
+
+{project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-02-success-criteria.md

package/_bmad/bme/_vortex/workflows/mvp/steps/step-02-success-criteria.md

@@ -0,0 +1,13 @@
+---
+step: 2
+workflow: mvp
+title: Define Success Criteria
+---
+
+# Step 2: Define Success Criteria
+
+How will you know if your MVP succeeded or failed?
+
+## Next Step
+
+{project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-03-smallest-test.md

package/_bmad/bme/_vortex/workflows/mvp/steps/step-03-smallest-test.md

@@ -0,0 +1,13 @@
+---
+step: 3
+workflow: mvp
+title: Design Smallest Test
+---
+
+# Step 3: Design Smallest Test
+
+What's the absolute minimum you can build to test your assumption?
+
+## Next Step
+
+{project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-04-scope-features.md

package/_bmad/bme/_vortex/workflows/mvp/steps/step-04-scope-features.md

@@ -0,0 +1,13 @@
+---
+step: 4
+workflow: mvp
+title: Scope MVP Features
+---
+
+# Step 4: Scope MVP Features
+
+What features are absolutely necessary? What can wait?
+
+## Next Step
+
+{project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-05-build-measure-learn.md

package/_bmad/bme/_vortex/workflows/mvp/steps/step-05-build-measure-learn.md

@@ -0,0 +1,13 @@
+---
+step: 5
+workflow: mvp
+title: Plan Build-Measure-Learn
+---
+
+# Step 5: Plan Build-Measure-Learn
+
+How will you build, measure, and learn from this MVP?
+
+## Next Step
+
+{project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-06-synthesize.md

package/_bmad/bme/_vortex/workflows/mvp/steps/step-06-synthesize.md

@@ -0,0 +1,28 @@
+---
+step: 6
+workflow: mvp
+title: Synthesize
+---
+
+# Step 6: Synthesize
+
+Create your final MVP specification.
+
+## Workflow Complete
+
+---
+
+## Vortex Compass
+
+Based on what you just completed, here are your evidence-driven options:
+
+| If you learned... | Consider next... | Agent | Why |
+|---|---|---|---|
+| MVP designed, ready to run | lean-experiment | Wade 🧪 | Execute the Build-Measure-Learn cycle |
+| Unsure about user needs | user-interview | Isla 🔍 | Validate user need before building |
+| MVP results available | learning-card | Max 🧭 | Capture what you learned from the MVP |
+
+> **Note:** These are evidence-based recommendations. You can navigate to any Vortex agent
+> at any time based on your judgment.
+
+**Or run Max's [VN] Vortex Navigation** for a full gap analysis across all streams.

package/_bmad/bme/_vortex/workflows/mvp/validate.md

@@ -0,0 +1,30 @@
+# Validate MVP Design
+
+**Status:** Coming in v1.2.0
+
+**Agent:** Wade (Lean Experiments Specialist)
+
+**Stream:** Externalize
+
+## Overview
+
+This validation workflow helps you review MVP designs to ensure they're truly minimal and focused on learning.
+
+## What Gets Validated
+
+- Is the value proposition clear and testable?
+- Are the riskiest assumptions identified?
+- Is the scope truly minimal (nothing extra)?
+- Are success metrics actionable (not vanity metrics)?
+- Is the build plan realistic?
+- Does this focus on learning over perfection?
+
+## Coming in v1.2.0
+
+This validation workflow will be available alongside the mvp workflow in March 2026.
+
+## Questions?
+
+For questions or to request early access:
+- GitHub Issues: https://github.com/amalik/convoke-agents/issues
+- Tag with: `workflow:mvp` and `v1.2.0`

package/_bmad/bme/_vortex/workflows/mvp/workflow.md

@@ -0,0 +1,36 @@
+---
+workflow: mvp
+type: step-file
+description: Design Minimum Viable Product specifications using Build-Measure-Learn
+author: Wade (lean-experiments-specialist)
+version: 1.2.0
+---
+
+# Design MVP Workflow
+
+This workflow guides you through designing a Minimum Viable Product that tests your riskiest assumptions.
+
+## What is an MVP?
+
+An MVP is NOT a feature-light product. It's the smallest thing that tests your riskiest assumption and enables validated learning.
+
+## Steps Overview
+
+1. **Identify Riskiest Assumption** - What could kill this idea?
+2. **Define Success Criteria** - How will you know if it worked?
+3. **Design Smallest Test** - What's the minimum to test the assumption?
+4. **Scope MVP Features** - What's absolutely necessary?
+5. **Plan Build-Measure-Learn** - How will you learn?
+6. **Synthesize** - Create MVP specification document
+
+## Output
+
+**Artifact:** MVP spec in `{output_folder}/mvp-spec-{mvp-name}-{date}.md`
+
+---
+
+## INITIALIZATION
+
+Load config from {project-root}/_bmad/bme/_vortex/config.yaml
+
+Load step: {project-root}/_bmad/bme/_vortex/workflows/mvp/steps/step-01-riskiest-assumption.md