convoke-agents 2.0.0

package/_bmad/bme/_vortex/workflows/signal-interpretation/steps/step-05-synthesize.md

@@ -0,0 +1,163 @@
+---
+step: 5
+workflow: signal-interpretation
+title: Synthesize & Route
+---
+
+# Step 5: Synthesize & Route
+
+Time to bring everything together. We've validated the experiment context, connected it to the production signal, described and analyzed the signal, and assessed anomalies and data quality. Now we produce the HC5 Signal Report artifact and route to the next step in the Vortex.
+
+## Why This Matters
+
+Production data is the most honest user feedback — it can't lie. But raw data isn't intelligence. Intelligence is data interpreted through context, connected to its experiment lineage, and packaged for the decision-maker. The HC5 artifact gives Max everything needed to make evidence-based decisions: what the signal is, where it came from, where it's heading, and what doesn't fit the expected model. No recommendations. No strategy. Just contextualized intelligence.
+
+## Your Task
+
+### 1. Review Your Signal Report
+
+Before we package everything, let's do a final quality pass on each section:
+
+**Signal Description:**
+
+| Field | Check |
+|-------|-------|
+| **Signal Summary** | Is it factual and one sentence? Could someone unfamiliar with the experiment understand what happened? |
+| **Signal Type** | Is the classification accurate? Does it match what was actually observed? |
+| **Severity** | Is the severity justified by the signal's deviation from experiment expectations? |
+| **Detection Method** | Is it clear how this signal came to attention? |
+| **Time Window** | Is the observation period precisely defined? |
+| **Affected Scope** | Are the affected users/segments/features clearly identified? |
+
+**Context (Experiment Lineage + Vortex History):**
+
+| Field | Check |
+|-------|-------|
+| **Experiment Lineage** | Can the signal be traced back to its originating experiment? Is the actual-vs-expected comparison clear? |
+| **Vortex History** | Are available upstream references (HC2, HC3, previous HC5) documented? |
+
+**Trend Analysis:**
+
+| Field | Check |
+|-------|-------|
+| **Trend Direction** | Does the classification reflect the data trajectory accurately? |
+| **Rate of Change** | Is it quantified with specific metrics, not vague descriptions? |
+| **Baseline Comparison** | Is the pre-experiment baseline explicitly stated? |
+| **Confidence** | Does the confidence level accurately reflect data quality? |
+
+**Anomaly Detection (if applicable):**
+
+| Field | Check |
+|-------|-------|
+| **Anomaly Description** | Is the unexpected behavior described factually, without speculative strategy? |
+| **Discovery Needed** | Is the HC10 routing decision documented with rationale? |
+
+**Data Quality:**
+
+| Field | Check |
+|-------|-------|
+| **Sample Size** | Is it sufficient for the conclusions drawn? |
+| **Confidence Level** | Does it honestly reflect data reliability? |
+
+### 2. Generate the HC5 Artifact
+
+I'll produce the HC5 Signal Report artifact with this structure:
+
+```yaml
+---
+contract: HC5
+type: artifact
+source_agent: noah
+source_workflow: signal-interpretation
+target_agents: [max]
+input_artifacts:
+  - path: "_bmad-output/vortex-artifacts/{your-hc4-artifact}"
+    contract: HC4
+created: YYYY-MM-DD
+---
+```
+
+**HC5 Required Body Sections:**
+1. **Signal Description** — Signal Summary, Signal Type, Severity, Detection Method, Time Window, Affected Scope
+2. **Context** — Experiment Lineage (Originating Experiment, Original Hypothesis, Experiment Outcome, Expected Production Behavior, Actual vs Expected) + Vortex History (Problem Definition, Hypothesis Origin, Previous Signals, Related Experiments)
+3. **Trend Analysis** — Trend Direction, Trend Duration, Rate of Change, Baseline Comparison, Confidence
+4. **Anomaly Detection** (when unexpected patterns detected) — Anomaly Description, Deviation from Expected, Potential Explanations, Discovery Needed, Discovery Focus
+5. **Data Quality** — Sample Size, Data Completeness, Known Biases, Confidence Level
+
+**This artifact explicitly does NOT include:**
+- Strategic recommendations (that is Max's domain)
+- Pivot/patch/persevere decisions (that is Max's domain)
+- Experiment design suggestions (that is Liam/Wade's domain)
+- Resource allocation recommendations (that is Max's domain)
+
+Noah produces intelligence — contextual, evidence-based signal interpretation. Max produces decisions.
+
+**Save to:** `{output_folder}/vortex-artifacts/hc5-signal-report-{date}.md`
+
+I'll create this file with all the sections above once you confirm the content is ready.
+
+### 3. Validation Questions
+
+Before we finalize, let's validate:
+
+**Evidence Check:**
+- [ ] Is every signal description grounded in observed data, not assumptions?
+- [ ] Can we trace the signal back to its originating experiment through the Experiment Lineage section?
+- [ ] Is the trend analysis based on actual metrics with quantified rates of change?
+
+**Completeness Check:**
+- [ ] Does the Signal Description include all 6 required fields?
+- [ ] Does the Context section include Experiment Lineage with all 5 required fields?
+- [ ] Does the Trend Analysis include all 5 required fields?
+- [ ] Does the Data Quality section include all required fields?
+- [ ] If anomalies were detected, does Anomaly Detection include all required fields?
+
+**Intelligence-Not-Strategy Check:**
+- [ ] Does the report contain zero strategic recommendations?
+- [ ] Does it avoid prescribing what to do about the signal?
+- [ ] Does it present findings in signal + context + trend format, leaving decisions to Max?
+- [ ] Would Max have everything needed to make an informed decision from this report?
+
+---
+
+## Your Turn
+
+Review the signal report sections. Confirm when you're ready for me to generate the final HC5 artifact.
+
+---
+
+**[a]** Advanced Elicitation — Deep dive into HC5 refinement with guided questioning
+**[p]** Party Mode — Bring in other Vortex agents for collaborative artifact critique
+**[c]** Continue — Generate the HC5 artifact and proceed to routing
+
+---
+
+## Vortex Compass
+
+Based on what you just completed, here are your evidence-driven options:
+
+| If you learned... | Consider next... | Agent | Why |
+|---|---|---|---|
+| Signal report complete with contextualized intelligence | learning-card | Max 🧭 | Signal report ready for decision (HC5) |
+| ⚡ Anomalous behavior detected not covered by original hypothesis | user-discovery | Isla 🔍 | Unexpected behavior warrants discovery research (HC10) |
+
+> **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.
+
+### ⚠️ Insufficient Evidence for Routing
+
+If the evidence gathered so far doesn't clearly point to a single next step:
+
+| To route to... | You need... |
+|----------------|-------------|
+| Max 🧭 | Complete HC5 signal report with all required sections and sufficient data quality confidence |
+| Isla 🔍 | Specific anomalous behavior identified with clear deviation from experiment expectations |
+
+**Workflow-specific signals:**
+- Signal description too vague → consider revisiting **step-03** for sharper analysis
+- Cannot connect signal to experiment context → consider revisiting **step-02** for deeper context loading
+- Data quality insufficient for any conclusion → gather more production data before proceeding
+
+**Recommended:** Revisit earlier steps to strengthen your signal report, or run **Max's [VN] Vortex Navigation** for a full gap analysis.

package/_bmad/bme/_vortex/workflows/signal-interpretation/workflow.md

@@ -0,0 +1,52 @@
+---
+workflow: signal-interpretation
+type: step-file
+description: Interpret production signals through experiment lineage and Vortex history to produce contextualized intelligence reports
+author: Noah (production-intelligence-specialist)
+version: 1.6.0
+---
+
+# Signal Interpretation Workflow
+
+This workflow guides you through interpreting a production signal by connecting it to its originating experiment context, analyzing trends, detecting anomalies, and producing an HC5 signal report for Max.
+
+## What is Signal Interpretation?
+
+Signal interpretation is the practice of reading production data through contextual lenses — not as isolated numbers on a dashboard, but as signals connected to the experiments that shaped them.
+
+Raw metrics mean nothing without interpretation frames. A 15% drop in engagement could be catastrophic or expected, depending on what experiment produced the current state, what hypothesis was tested, and what production behavior was predicted. The signal indicates one thing when viewed in isolation and something entirely different when viewed through experiment lineage.
+
+This workflow connects each production signal to its originating experiment, analyzes the trend in context, surfaces anomalies that dashboards hide, and packages everything as intelligence for the decision-maker — without prescribing what to do about it. Here's what we're seeing in context: that's the deliverable. Not strategy. Not recommendations. Contextualized intelligence.
+
+## 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. **Setup & Input Validation** - Validate experiment context input (HC4 artifact or equivalent)
+2. **Experiment Context & Signal Connection** - Load experiment lineage and connect it to the production signal
+3. **Signal Description & Trend Analysis** - Describe the signal factually and analyze its trajectory against baselines
+4. **Anomaly Detection & Data Quality** - Surface unexpected behavior patterns and assess data reliability
+5. **Synthesize & Route** - Produce HC5 signal report artifact and route via Compass
+
+## Output
+
+**Artifact:** HC5 Signal Report markdown file in `{output_folder}/vortex-artifacts/hc5-signal-report-{date}.md`
+
+**Template:** None (HC5 artifact is generated inline during Step 5)
+
+**Schema:** Conforms to HC5 contract (`_bmad/bme/_vortex/contracts/hc5-signal-report.md`)
+
+**Consumer:** Max (learning-card) uses this to make evidence-based decisions grounded in production reality.
+
+---
+
+## INITIALIZATION
+
+Load config from {project-root}/_bmad/bme/_vortex/config.yaml
+
+Load step: {project-root}/_bmad/bme/_vortex/workflows/signal-interpretation/steps/step-01-setup.md

package/_bmad/bme/_vortex/workflows/user-discovery/steps/step-01-discovery-scope.md

@@ -0,0 +1,77 @@
+---
+step: 1
+workflow: user-discovery
+title: Define Discovery Scope
+---
+
+# Step 1: Define Discovery Scope
+
+Before any research begins, you need to be explicit about what you are trying to learn and why it matters. A well-scoped discovery prevents the two most common research failures: learning everything about nothing, and learning nothing about everything.
+
+## Why This Matters
+
+Discovery research is expensive in time, attention, and participant goodwill. A clear scope ensures every research activity produces knowledge that moves your product forward. Without it, teams collect volumes of data and still cannot answer the questions that matter.
+
+## Your Task
+
+### 1. State the Discovery Goal
+
+What decision or direction will this research inform? Be specific about the business or product context that makes this discovery necessary right now.
+
+**Strong examples:**
+- "We're entering the small-business accounting market and need to understand how owners currently manage their finances before we design our onboarding flow."
+- "Churn increased 15% last quarter among enterprise accounts. We need to discover what changed in their workflows or expectations."
+- "We have three feature directions for Q3. We need to understand which user problems are most urgent to decide where to invest."
+
+**Weak examples:**
+- "We want to understand our users better." (Too vague -- what would change if you did?)
+- "We need to do some research." (Not a goal, it's an activity.)
+
+### 2. Formulate Research Questions (3-7)
+
+Research questions are NOT interview questions. They are the knowledge gaps you need to close. Good research questions share three traits: they are specific, answerable through observation or conversation, and relevant to a decision you face.
+
+**Example research questions:**
+- "What does a typical day look like for a small-business owner managing finances? What tools, workarounds, and people are involved?"
+- "At what point in the workflow do enterprise users become frustrated enough to consider alternatives?"
+- "What criteria do users actually use when choosing between competing tools, versus what they say they use?"
+- "How do environmental factors (noise, interruptions, device switching) affect how users interact with our product?"
+
+**Tip:** If a question can be answered with a simple yes/no, it is not a research question -- it is a hypothesis. Save those for experiments (Wade can help with that).
+
+### 3. Define the User Population
+
+Who do you need to learn about? Be specific enough to recruit effectively, but broad enough that you do not accidentally study only your happiest customers.
+
+**Consider:**
+- Current users vs. non-users vs. former users
+- Power users vs. casual users vs. struggling users
+- Different roles, contexts, or segments
+- People who might use your product but currently solve the problem differently
+
+### 4. Set Success Criteria for Discovery
+
+How will you know when discovery is "done enough"? Perfect knowledge is impossible, but you should define a threshold.
+
+**Examples:**
+- "We can confidently describe 3 distinct user segments and their primary workflows."
+- "We have evidence-backed answers to all 5 research questions from at least 3 independent sources each."
+- "We've identified the top 3 unmet needs and can rank them by severity with supporting evidence."
+
+---
+
+## Your Turn
+
+Please provide:
+1. Your discovery goal (what decision will this research inform?)
+2. 3-7 research questions you need to answer
+3. The user population you plan to study
+4. Your success criteria for completing discovery
+
+I'll review your scope for clarity, specificity, and feasibility before we proceed.
+
+## Next Step
+
+When your discovery scope is defined and confirmed, I'll load:
+
+{project-root}/_bmad/bme/_vortex/workflows/user-discovery/steps/step-02-research-methods.md

package/_bmad/bme/_vortex/workflows/user-discovery/steps/step-02-research-methods.md

@@ -0,0 +1,152 @@
+---
+step: 2
+workflow: user-discovery
+title: Choose Research Methods
+---
+
+# Step 2: Choose Research Methods
+
+Now that you know what you need to learn, let's choose the right methods to learn it. Different questions call for different approaches, and combining methods produces richer, more trustworthy findings.
+
+## Why This Matters
+
+Relying on a single research method creates blind spots. Interviews reveal what people say but not always what they do. Analytics show behavior but not motivation. Observation captures context but is hard to scale. The best discovery plans triangulate -- using multiple methods to validate findings from different angles.
+
+## Method Menu
+
+### Contextual Inquiry
+
+**What it is:** Observing and interviewing users in their actual environment while they perform real tasks.
+
+**Best for answering:**
+- "What does their actual workflow look like?" (not the idealized version)
+- "What environmental factors affect their experience?"
+- "What workarounds and tools do they combine to get things done?"
+
+**Practical guidance:**
+- Visit users where they work (office, home, field)
+- Ask them to perform their normal tasks while you observe
+- Use the "master-apprentice" model: they are the expert, you are learning
+- Capture the physical environment (desk setup, sticky notes, multiple monitors)
+- Session length: 60-90 minutes per participant
+- Recommended sample: 5-8 participants per segment
+
+**Watch out for:** The Hawthorne effect -- people change behavior when observed. Schedule enough time that participants forget you're there.
+
+### Diary Studies
+
+**What it is:** Participants log their experiences over time (days or weeks) in a structured format you provide.
+
+**Best for answering:**
+- "How does this experience unfold over time?"
+- "What triggers engagement or frustration?"
+- "How frequently do specific problems actually occur?"
+
+**Practical guidance:**
+- Provide a simple logging template (time, activity, emotion, notes)
+- Use tools participants already have (text messages, photos, simple forms)
+- Keep the burden low: 2-3 entries per day, 5 minutes each
+- Duration: 1-4 weeks depending on the cycle you're studying
+- Send daily reminders and respond to entries to maintain engagement
+- Recommended sample: 8-15 participants
+
+**Watch out for:** Participant fatigue. Front-load the most critical days and keep the logging format minimal.
+
+### Ethnographic Observation
+
+**What it is:** Immersive observation of users in their natural context without direct interaction or task direction.
+
+**Best for answering:**
+- "What do users actually do versus what they report doing?"
+- "What social and environmental dynamics influence behavior?"
+- "What implicit norms or habits exist that users cannot articulate?"
+
+**Practical guidance:**
+- Observe without intervening or asking questions during the session
+- Take detailed field notes: what you see, hear, and the environment
+- Note timestamps so you can reconstruct sequences later
+- Photograph or sketch the physical environment (with permission)
+- Debrief with participants afterward to clarify what you observed
+- Session length: 2-4 hours per observation
+- Recommended sample: 4-6 sessions per context
+
+**Watch out for:** Observer bias -- you will see what you expect to see. Bring a second observer when possible and compare notes independently before discussing.
+
+### Surveys
+
+**What it is:** Structured questionnaires distributed to a larger sample for quantitative and qualitative data.
+
+**Best for answering:**
+- "How prevalent is this behavior or attitude across our user base?"
+- "What are the relative priorities among known needs?"
+- "How do different segments compare on key dimensions?"
+
+**Practical guidance:**
+- Use surveys to validate and quantify patterns found in qualitative research, not to discover new patterns
+- Keep surveys under 15 questions and under 5 minutes completion time
+- Mix closed-ended (scales, multiple choice) with 1-2 open-ended questions
+- Pilot test with 3-5 people before full distribution
+- Target a minimum of 30 responses per segment for meaningful patterns
+- Recommended sample: 50-200+ depending on segment count
+
+**Watch out for:** Leading questions, double-barreled questions, and social desirability bias. Have someone outside the team review the survey before distribution.
+
+### Analytics Review
+
+**What it is:** Analyzing existing behavioral data from your product, website, or other systems.
+
+**Best for answering:**
+- "What do users actually do in our product?"
+- "Where do they drop off, get stuck, or abandon?"
+- "Which features are used most and least?"
+- "How do behavioral patterns differ across segments?"
+
+**Practical guidance:**
+- Start with the user journey and map data to each stage
+- Look for anomalies: unexpected drop-offs, surprising usage patterns, features with high engagement but low satisfaction
+- Combine quantitative metrics with qualitative context (why is this number what it is?)
+- Create cohort analyses to compare segments
+- Time investment: 4-8 hours for a thorough review
+
+**Watch out for:** Analytics show what happened but not why. Never draw motivational conclusions from behavioral data alone.
+
+## Choosing Your Methods
+
+### Method Selection Matrix
+
+| Research Question | Contextual Inquiry | Diary Study | Observation | Survey | Analytics |
+|---|---|---|---|---|---|
+| Understand workflows | Strong | Medium | Strong | Weak | Medium |
+| Discover motivations | Strong | Strong | Weak | Medium | Weak |
+| Measure prevalence | Weak | Weak | Weak | Strong | Strong |
+| Track over time | Weak | Strong | Weak | Medium | Strong |
+| Uncover environment | Strong | Medium | Strong | Weak | Weak |
+| Validate patterns | Medium | Medium | Medium | Strong | Strong |
+
+### Recommended Combinations
+
+**Deep understanding of a new domain:** Contextual Inquiry + Diary Studies
+**Validating known problems at scale:** Analytics Review + Survey
+**Comprehensive discovery:** Contextual Inquiry + Analytics Review + Survey (confirmation)
+**Longitudinal insight:** Diary Study + Analytics Review
+
+## Your Task
+
+For each of your research questions from Step 1:
+1. Select 1-2 methods that best answer that question
+2. Explain why you chose those methods
+3. Note any constraints (budget, time, access to users) that limit your options
+
+Aim for at least 2 different methods total across all questions to enable triangulation.
+
+---
+
+## Your Turn
+
+Please share your method selections for each research question. I'll review them for appropriateness and suggest adjustments if needed.
+
+## Next Step
+
+When your methods are selected and confirmed, I'll load:
+
+{project-root}/_bmad/bme/_vortex/workflows/user-discovery/steps/step-03-research-plan.md

package/_bmad/bme/_vortex/workflows/user-discovery/steps/step-03-research-plan.md

@@ -0,0 +1,159 @@
+---
+step: 3
+workflow: user-discovery
+title: Plan Research Activities
+---
+
+# Step 3: Plan Research Activities
+
+You know what to learn and which methods to use. Now let's build a concrete plan: who participates, when things happen, what materials you need, and how you handle ethics.
+
+## Why This Matters
+
+Research without a plan produces scattered, unreliable data. A clear plan ensures you recruit the right participants, prepare the right materials, and respect the people who give you their time and trust. It also gives stakeholders confidence that the investment will pay off.
+
+## Your Task
+
+### 1. Participant Recruitment
+
+**How many participants do you need?**
+
+Use these guidelines based on your chosen methods:
+- Contextual inquiry: 5-8 per segment
+- Diary studies: 8-15 total (expect 20-30% dropout)
+- Ethnographic observation: 4-6 sessions
+- Surveys: 50+ per segment (30 minimum for any pattern)
+- Analytics review: N/A (uses existing data)
+
+**Where will you find participants?**
+
+| Source | Pros | Cons |
+|--------|------|------|
+| Existing users | Easy access, real context | Survivorship bias |
+| Former users | Understand churn | Harder to reach, may be negative |
+| Non-users | Fresh perspective | May not understand domain |
+| Recruiting panels | Fast, screened | May not match real users |
+| Community forums | Self-selected, passionate | Skewed toward vocal minority |
+| Partner referrals | Trusted introduction | Small sample, filtered |
+
+**Screening criteria:**
+- What characteristics MUST participants have? (Include criteria)
+- What characteristics DISQUALIFY someone? (Exclude criteria)
+- How will you ensure diversity within your target population?
+
+**Example screening questions:**
+- "How often do you [relevant activity]?" (filters for frequency)
+- "What tools do you currently use for [task]?" (filters for context)
+- "How would you describe your experience level with [domain]?" (filters for expertise)
+
+### 2. Timeline
+
+Build a realistic timeline that accounts for the reality that recruitment always takes longer than expected.
+
+**Template timeline:**
+
+| Week | Activity | Deliverable |
+|------|----------|-------------|
+| 1 | Finalize plan, create materials, begin recruitment | Screener, guides, consent forms |
+| 2 | Recruitment and pilot sessions (1-2) | Refined guides based on pilot |
+| 3-4 | Core research sessions | Raw notes, recordings, artifacts |
+| 5 | Data organization and initial synthesis | Organized data set, initial themes |
+| 6 | Final synthesis and report | Discovery report artifact |
+
+Adjust based on your methods. Diary studies extend the core research phase. Analytics reviews can compress the timeline significantly.
+
+### 3. Research Materials
+
+**For each method, prepare:**
+
+**Contextual Inquiry:**
+- Discussion guide (topics, not rigid questions)
+- Observation checklist (what to look for in the environment)
+- Note-taking template (timestamp, observation, interpretation columns)
+- Recording equipment and backup plan
+
+**Diary Studies:**
+- Participant briefing document (what to log, how often, for how long)
+- Logging template or tool setup
+- Reminder schedule
+- Incentive structure (when and how participants are compensated)
+
+**Ethnographic Observation:**
+- Observation protocol (where to position, what to record)
+- Field notes template
+- Camera/sketch supplies for environment capture
+- Post-observation debrief questions
+
+**Surveys:**
+- Draft questionnaire with all questions
+- Pilot test plan (3-5 people review before launch)
+- Distribution channel and timeline
+- Response target and follow-up plan
+
+**Analytics Review:**
+- Data sources and access confirmed
+- Key metrics and events to analyze
+- Cohort definitions
+- Analysis framework (what comparisons to make)
+
+### 4. Ethical Considerations
+
+Research involves real people. Handle their trust with care.
+
+**Informed Consent:**
+- Explain the purpose of the research in plain language
+- Describe what participation involves (time, activities, recording)
+- State that participation is voluntary and can be stopped at any time
+- Explain how data will be stored, used, and eventually deleted
+- Get written consent before any session begins
+
+**Privacy and Data Protection:**
+- Anonymize data as early as possible (use participant codes, not names)
+- Store recordings and notes securely (encrypted, access-limited)
+- Define a data retention period and stick to it
+- Never share raw data outside the research team without consent
+- Be especially careful with screenshots, photos, and screen recordings
+
+**Compensation:**
+- Compensate participants fairly for their time
+- Pay regardless of whether they complete the full session
+- Common rates: $50-150/hour for professionals, gift cards for consumers
+- For diary studies: milestone payments to maintain engagement
+
+**Vulnerability:**
+- Be aware that research can surface sensitive topics
+- Have a plan for if a participant becomes distressed
+- Never pressure someone to continue if they seem uncomfortable
+- Researchers are not therapists -- know your boundaries
+
+### 5. Logistics Checklist
+
+- [ ] Recruitment screener created and distributed
+- [ ] Consent forms prepared
+- [ ] Discussion guides / observation protocols drafted
+- [ ] Recording equipment tested
+- [ ] Note-taking templates ready
+- [ ] Incentives budgeted and procured
+- [ ] Calendar invites sent with clear instructions
+- [ ] Backup plan for no-shows (over-recruit by 20%)
+- [ ] Data storage location set up and access confirmed
+- [ ] Pilot session scheduled (at least 1 before core research)
+
+---
+
+## Your Turn
+
+Please provide:
+1. Your recruitment plan (how many, from where, screening criteria)
+2. Your timeline (adapt the template above to your reality)
+3. Which materials you need to prepare for your chosen methods
+4. Your approach to consent, privacy, and compensation
+5. Any logistical concerns or constraints
+
+I'll review your plan for feasibility and completeness.
+
+## Next Step
+
+When your research plan is confirmed, I'll load:
+
+{project-root}/_bmad/bme/_vortex/workflows/user-discovery/steps/step-04-execute.md