nemonix-cli 1.0.0

package/src/config/claude-code-config/skills/workflow-brainstorm/steps/step-02-challenge.md

@@ -0,0 +1,238 @@
+---
+name: step-02-challenge
+description: Critical challenge phase - devil's advocate stress-testing all findings
+next_step: steps/step-03-synthesize.md
+---
+
+# Phase 2: Critical Challenge
+
+**Role: DEVIL'S ADVOCATE** - Question everything mercilessly
+
+---
+
+<available_state>
+From previous steps:
+
+- `{topic}` - The research topic
+- `{economy_mode}` - If true, use direct tool calls instead of subagents
+- `{fast_mode}` - If true, this step was SKIPPED (you shouldn't be here)
+- `{save_file}` - If true, save session to file
+- `{session_path}` - Path to session file (if saving)
+- `{phase_1_discoveries}` - Broad findings from Phase 1
+</available_state>
+
+---
+
+<mandatory_rules>
+## MANDATORY EXECUTION RULES (READ FIRST)
+
+- 🔥 CHALLENGE every finding from Phase 1 - nothing is sacred
+- ❓ SEARCH for contradicting evidence and opposing viewpoints
+- 🕳️ IDENTIFY gaps, assumptions, and weak points
+- 🧠 UPDATE mental model based on challenges - change your mind!
+- 🚫 FORBIDDEN: Accepting anything at face value
+- 🚫 FORBIDDEN: Defending Phase 1 findings against legitimate challenges
+</mandatory_rules>
+
+---
+
+<execution_sequence>
+
+<review_findings>
+**1. Review Phase 1 with Skeptical Eyes**
+
+Look at `{phase_1_discoveries}` and ask for EACH finding:
+- What evidence supports this? Is it strong or weak?
+- Who benefits from this being true? (follow the incentives)
+- What's the sample size / scope of this claim?
+- Is this recency bias? survivorship bias? selection bias?
+- What would need to be true for this to be wrong?
+
+**Mark each finding:**
+- ✅ Strong evidence (multiple independent sources)
+- ⚠️ Moderate evidence (some support, some gaps)
+- ❌ Weak evidence (assumption, anecdote, or single source)
+</review_findings>
+
+<challenge_agents>
+**2. Execute Challenge Research**
+
+<economy_mode_research>
+**If `{economy_mode}` = true:** Use direct tool calls
+
+**Web Search** (use WebSearch tool directly)
+- Search: `{topic} problems failures criticisms`
+- Search: `{topic} downsides risks hidden costs`
+- Search: `{topic} alternatives why not to use`
+
+Focus on finding counter-arguments and failures.
+</economy_mode_research>
+
+<standard_mode_research>
+**If `{economy_mode}` = false:** Launch parallel subagents
+
+**CRITICAL: Launch ALL agents in a SINGLE message for parallel execution.**
+
+**Agent 1: Devil's Advocate Search** (`websearch`)
+```
+Search for criticisms and problems with: {topic} + [popular approaches from Phase 1]
+
+Find:
+1. Criticisms and failures of the popular approaches
+2. Failed implementations and why they failed
+3. Experts who disagree with the mainstream view
+4. Hidden costs, risks, or downsides
+5. Cautionary tales and horror stories
+
+Focus on finding the OTHER side of the story.
+```
+
+**Agent 2: Gap Analysis Search** (`websearch`)
+```
+Search for what's missing from discussions about: {topic}
+
+Find:
+1. Edge cases rarely discussed
+2. Long-term consequences often ignored
+3. Alternative approaches that get overlooked
+4. Questions that experts avoid answering
+5. Assumptions that are rarely questioned
+
+Focus on what's NOT being said.
+```
+</standard_mode_research>
+</challenge_agents>
+
+<systematic_challenge>
+**3. Challenge Each Phase 1 Finding**
+
+For each major finding from Phase 1:
+
+**Finding: "[Phase 1 finding]"**
+
+**Challenge Questions:**
+- What evidence contradicts this?
+- Who disagrees and why might they be right?
+- What could go wrong if we act on this belief?
+- What's the weakest link in the reasoning?
+- What assumption is this built on?
+
+**Counter-evidence found:**
+- [Contradicting source/evidence]
+
+**Updated belief:**
+- Original confidence: [High/Medium/Low]
+- New confidence: [Higher/Same/Lower]
+- Reason for change: [What new info emerged]
+
+Repeat for each major finding.
+</systematic_challenge>
+
+<gap_analysis>
+**4. Identify Gaps and Assumptions**
+
+**Unanswered Questions:**
+- [Critical question that Phase 1 didn't answer]
+- [Question that emerged from challenges]
+
+**Hidden Assumptions:**
+- We're assuming [assumption 1] - but is this valid?
+- We're assuming [assumption 2] - what if we're wrong?
+
+**Missing Evidence:**
+- We don't have data on [important aspect]
+- No one seems to address [gap]
+
+**Blind Spots:**
+- Phase 1 research may have missed [area]
+- We should have also looked at [unexplored angle]
+</gap_analysis>
+
+<synthesize_challenges>
+**5. Synthesize Challenge Findings**
+
+```
+## Phase 2 Challenges: {topic}
+
+**Findings That Survived Challenge:**
+- [Finding]: Still confident because [reason]
+
+**Findings Now In Doubt:**
+- [Finding]: Challenged by [evidence], new confidence: [level]
+
+**Completely Reversed:**
+- [Finding]: Was wrong because [reason]
+
+**Critical Gaps Identified:**
+1. [Gap 1] - This matters because...
+2. [Gap 2] - We can't proceed without knowing...
+
+**Assumptions We're Making:**
+1. [Assumption] - Risk if wrong: [consequence]
+
+**Updated Mental Model:**
+After challenging Phase 1 findings, my view has shifted:
+- [What changed]
+- [What stayed the same but is now more confident]
+- [What needs more investigation]
+```
+
+Store in `{phase_2_challenges}`
+</synthesize_challenges>
+
+<save_progress>
+**6. Save Progress** (if `{save_file}` is true)
+
+Append to `{session_path}`:
+```markdown
+## Phase 2: Critical Challenge
+
+[Phase 2 challenges content]
+
+---
+```
+</save_progress>
+
+</execution_sequence>
+
+---
+
+<success_metrics>
+
+- Every Phase 1 finding actively challenged
+- Counter-evidence actively searched for
+- Assumptions explicitly identified
+- Mental model updated based on challenges
+- Confidence levels adjusted honestly
+- Gaps and blind spots documented
+</success_metrics>
+
+<failure_modes>
+
+- Defending Phase 1 findings instead of challenging them
+- Only seeking weak counter-arguments to dismiss
+- Not actually changing beliefs based on evidence
+- Skipping the gap analysis
+- Being skeptical in word but not in action
+</failure_modes>
+
+---
+
+<next_step_directive>
+**CRITICAL:** After presenting Phase 2 challenges, immediately state:
+
+"**Phase 2 complete.** We've stress-tested our findings and updated our mental model.
+
+**What changed:**
+- [Key shift in understanding]
+- [Finding that was strengthened]
+- [Finding that was weakened or reversed]
+
+**NEXT: Phase 3 - Multi-Lens Synthesis**
+
+Now I become the MULTI-LENS SYNTHESIZER. We'll examine everything through 5 different expert perspectives to find patterns and build a complete picture.
+
+Loading `step-03-synthesize.md`..."
+
+Then load and execute `steps/step-03-synthesize.md`.
+</next_step_directive>

package/src/config/claude-code-config/skills/workflow-brainstorm/steps/step-03-synthesize.md

@@ -0,0 +1,337 @@
+---
+name: step-03-synthesize
+description: Multi-lens synthesis phase - analyze from 5 expert perspectives
+next_step: steps/step-04-action.md
+---
+
+# Phase 3: Multi-Lens Synthesis
+
+**Role: MULTI-LENS SYNTHESIZER** - See from every angle, find the patterns
+
+---
+
+<available_state>
+From previous steps:
+
+- `{topic}` - The research topic
+- `{economy_mode}` - If true, use direct tool calls instead of subagents
+- `{fast_mode}` - If true, use 3 perspectives instead of 5
+- `{save_file}` - If true, save session to file
+- `{session_path}` - Path to session file (if saving)
+- `{phase_1_discoveries}` - Broad findings from Phase 1
+- `{phase_2_challenges}` - Challenges from Phase 2 (empty if fast_mode skipped Phase 2)
+</available_state>
+
+---
+
+<mandatory_rules>
+## MANDATORY EXECUTION RULES (READ FIRST)
+
+- 🎭 ADOPT each perspective FULLY before moving to next
+- 🔗 FIND patterns that emerge across multiple perspectives
+- ⚖️ WEIGHT insights by strength of reasoning, not by popularity
+- 🔍 DO targeted research for each perspective if needed
+- 🚫 FORBIDDEN: Staying in one perspective too long
+- 🚫 FORBIDDEN: Skipping perspectives that feel uncomfortable
+</mandatory_rules>
+
+---
+
+<the_five_lenses>
+## Expert Perspectives
+
+**Standard mode (5 perspectives):**
+| Lens | Character | Core Question |
+|------|-----------|---------------|
+| 1 | The Pragmatist | What actually works in the real world? |
+| 2 | The Perfectionist | What's the ideal solution if constraints didn't exist? |
+| 3 | The Skeptic | What could go terribly wrong? |
+| 4 | The Expert | What would a 10-year veteran prioritize? |
+| 5 | The Beginner | What obvious questions are we not asking? |
+
+**Fast mode (3 perspectives):** Pragmatist, Skeptic, Expert only
+</the_five_lenses>
+
+---
+
+<execution_sequence>
+
+<prepare>
+**1. Prepare for Multi-Perspective Analysis**
+
+Review available findings to ground the analysis:
+
+**If `{fast_mode}` = true:**
+"With Phase 1 complete (Phase 2 skipped for speed), we have:
+- **Broad landscape** from exploration
+
+Now I'll analyze `{topic}` through 3 expert lenses (Pragmatist, Skeptic, Expert)."
+
+**If `{fast_mode}` = false:**
+"With Phases 1 and 2 complete, we have:
+- **Broad landscape** from exploration
+- **Stress-tested findings** from challenges
+- **Updated beliefs** based on evidence
+
+Now I'll analyze `{topic}` through 5 expert lenses to build a complete picture."
+</prepare>
+
+<lens_1_pragmatist>
+**2. LENS 1: The Pragmatist**
+
+*"Theory is nice, but what actually works in production?"*
+
+**Targeted research** (quick websearch if needed):
+- Real-world case studies and post-mortems
+- Production experiences from practitioners
+
+**Analysis:**
+```
+## The Pragmatist's View
+
+**What actually works:**
+- [Approach that has proven track record]
+- [Pattern that teams consistently succeed with]
+
+**What looks good but fails in practice:**
+- [Approach that sounds good but has hidden issues]
+
+**The 80/20:**
+- 80% of value comes from: [simple approach]
+- Avoid premature optimization of: [complex approach]
+
+**Pragmatist's recommendation:**
+"If I had to ship this tomorrow, I would..."
+```
+</lens_1_pragmatist>
+
+<lens_2_perfectionist>
+**3. LENS 2: The Perfectionist** ⏭️ *Skip if `{fast_mode}` = true*
+
+*"If we had unlimited time and resources, what's the ideal?"*
+
+**Analysis:**
+```
+## The Perfectionist's View
+
+**The ideal solution:**
+- [Theoretically perfect approach]
+- [Why this is considered "best" in the abstract]
+
+**What we'd need to get there:**
+- [Resource 1]
+- [Capability 2]
+- [Time investment 3]
+
+**Is perfection worth pursuing?**
+- Gap between pragmatic and perfect: [assessment]
+- Value of closing that gap: [analysis]
+
+**Perfectionist's recommendation:**
+"The gold standard would be... but consider [tradeoffs]"
+```
+</lens_2_perfectionist>
+
+<lens_3_skeptic>
+**4. LENS 3: The Skeptic**
+
+*"What's the worst that could happen? What are the hidden risks?"*
+
+**Targeted research** (quick websearch if needed):
+- Security incidents, failures, vulnerabilities
+- Long-term consequences and technical debt
+
+**Analysis:**
+```
+## The Skeptic's View
+
+**Hidden risks no one talks about:**
+- [Risk 1]: Could happen if [condition]
+- [Risk 2]: Often overlooked because [reason]
+
+**Worst-case scenarios:**
+- If [approach] fails, the consequence is [impact]
+
+**Long-term concerns:**
+- Technical debt: [assessment]
+- Lock-in risks: [analysis]
+- Scalability cliffs: [prediction]
+
+**What could make this completely wrong:**
+- [Scenario that would invalidate the approach]
+
+**Skeptic's warning:**
+"Watch out for... and have a plan for..."
+```
+</lens_3_skeptic>
+
+<lens_4_expert>
+**5. LENS 4: The Expert**
+
+*"What would someone with 10 years of experience prioritize?"*
+
+**Analysis:**
+```
+## The Expert's View
+
+**What experts actually focus on:**
+- [Priority 1]: Because experience shows...
+- [Priority 2]: This is what separates success from failure
+
+**Common mistakes of those newer to this:**
+- [Mistake 1]: Seems right but leads to...
+- [Mistake 2]: Overlooking...
+
+**The non-obvious insight:**
+- [Counterintuitive wisdom that comes from experience]
+
+**Expert's prioritization:**
+1. [Most important]: Because...
+2. [Second priority]: Then...
+3. [Can wait]: Worry about this later...
+
+**Expert's recommendation:**
+"If you only do one thing right, make it..."
+```
+</lens_4_expert>
+
+<lens_5_beginner>
+**6. LENS 5: The Beginner** ⏭️ *Skip if `{fast_mode}` = true*
+
+*"What basic questions are we not asking because we assume everyone knows?"*
+
+**Analysis:**
+```
+## The Beginner's View
+
+**Questions that feel "too basic" to ask:**
+- Why do we actually need [assumption]?
+- What if we just [simple alternative]?
+- Has anyone tried NOT doing [standard practice]?
+
+**Assumptions we're making:**
+- We assume [X] is necessary, but is it?
+- We assume [Y] is the goal, but who decided?
+
+**Fresh perspective:**
+- [Observation that experts overlook due to familiarity]
+
+**Beginner's challenge:**
+"Wait, why don't we just...?"
+```
+</lens_5_beginner>
+
+<pattern_recognition>
+**7. Find Patterns Across Perspectives**
+
+**Cross-Perspective Analysis:**
+
+**If `{fast_mode}` = true (3 perspectives):**
+```
+## Patterns Emerging (Fast Mode)
+
+**Convergence:**
+- [Insight that Pragmatist, Skeptic, and Expert all agree on]
+
+**Key tension:**
+- Pragmatist vs Skeptic: [tradeoff between speed and caution]
+
+**Synthesis:**
+The clearest signal across all 3 perspectives is...
+```
+
+**If `{fast_mode}` = false (5 perspectives):**
+```
+## Patterns Emerging
+
+**Convergence (all perspectives agree):**
+- [Insight that appears in 3+ perspectives]
+- [Consistent recommendation]
+
+**Tension (perspectives conflict):**
+- Pragmatist vs Perfectionist: [tradeoff]
+- Expert vs Beginner: [assumption being challenged]
+
+**Strongest signal:**
+When [multiple perspectives] all point to [conclusion], that's our strongest insight.
+
+**Surprising emergence:**
+The Beginner's question about [X] actually exposed a blind spot in the Expert's view about [Y].
+
+**Synthesis:**
+Weighing all perspectives, the clearest thread is...
+```
+
+Store in `{phase_3_synthesis}`
+</pattern_recognition>
+
+<save_progress>
+**8. Save Progress** (if `{save_file}` is true)
+
+Append to `{session_path}`:
+```markdown
+## Phase 3: Multi-Lens Synthesis
+
+### The Pragmatist
+[Pragmatist analysis]
+
+### The Perfectionist
+[Perfectionist analysis]
+
+### The Skeptic
+[Skeptic analysis]
+
+### The Expert
+[Expert analysis]
+
+### The Beginner
+[Beginner analysis]
+
+### Patterns Emerging
+[Cross-perspective synthesis]
+
+---
+```
+</save_progress>
+
+</execution_sequence>
+
+---
+
+<success_metrics>
+
+- All perspectives explored (5 standard, 3 if fast_mode)
+- Each lens given genuine consideration
+- Targeted research done where perspectives needed data
+- Patterns and tensions identified across perspectives
+- Synthesis captures the strongest emergent insights
+</success_metrics>
+
+<failure_modes>
+
+- Giving lip service to perspectives without genuine analysis
+- Skipping the Skeptic because findings are uncomfortable
+- Letting one perspective dominate the synthesis
+- Not finding the cross-perspective patterns
+- Treating perspectives as checklist rather than genuine lenses
+</failure_modes>
+
+---
+
+<next_step_directive>
+**CRITICAL:** After presenting multi-lens synthesis, immediately state:
+
+"**Phase 3 complete.** We've seen `{topic}` from 5 expert perspectives and found the patterns.
+
+**Strongest insights:**
+- [Key convergence across perspectives]
+- [Important tension to navigate]
+
+**NEXT: Phase 4 - Action Crystallization**
+
+Now I become the STRATEGIC ADVISOR. Time to crystallize everything into clear recommendations, honest confidence levels, and actionable next steps.
+
+Loading `step-04-action.md`..."
+
+Then load and execute `steps/step-04-action.md`.
+</next_step_directive>