reasoning-dialectical
Synthesize competing positions through structured thesis-antithesis-synthesis process. Use when stakeholders disagree, trade-offs exist, or multiple valid perspectives need integration. Produces integrated positions with acknowledged trade-offs.
Best use case
reasoning-dialectical is best used when you need a repeatable AI agent workflow instead of a one-off prompt. It is especially useful for teams working in multi. Synthesize competing positions through structured thesis-antithesis-synthesis process. Use when stakeholders disagree, trade-offs exist, or multiple valid perspectives need integration. Produces integrated positions with acknowledged trade-offs.
Synthesize competing positions through structured thesis-antithesis-synthesis process. Use when stakeholders disagree, trade-offs exist, or multiple valid perspectives need integration. Produces integrated positions with acknowledged trade-offs.
Users should expect a more consistent workflow output, faster repeated execution, and less time spent rewriting prompts from scratch.
Practical example
Example input
Use the "reasoning-dialectical" skill to help with this workflow task. Context: Synthesize competing positions through structured thesis-antithesis-synthesis process. Use when stakeholders disagree, trade-offs exist, or multiple valid perspectives need integration. Produces integrated positions with acknowledged trade-offs.
Example output
A structured workflow result with clearer steps, more consistent formatting, and an output that is easier to reuse in the next run.
When to use this skill
- Use this skill when you want a reusable workflow rather than writing the same prompt again and again.
When not to use this skill
- Do not use this when you only need a one-off answer and do not need a reusable workflow.
- Do not use it if you cannot install or maintain the related files, repository context, or supporting tools.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/reasoning-dialectical/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How reasoning-dialectical Compares
| Feature / Agent | reasoning-dialectical | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Synthesize competing positions through structured thesis-antithesis-synthesis process. Use when stakeholders disagree, trade-offs exist, or multiple valid perspectives need integration. Produces integrated positions with acknowledged trade-offs.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Dialectical Reasoning
Synthesize opposing views into higher-order resolution. The logic of productive disagreement.
## Type Signature
```
Dialectical : Thesis → Antithesis → Synthesis
Where:
Thesis : Position × Evidence × Stakeholder → ArgumentA
Antithesis : ArgumentA → CounterPosition × Evidence × Stakeholder → ArgumentB
Synthesis : (ArgumentA, ArgumentB) → IntegratedPosition × Tradeoffs
```
## When to Use
**Use dialectical when:**
- Stakeholders hold opposing valid positions
- Trade-offs need explicit analysis
- Strategic tension requires resolution
- Multiple perspectives each have merit
- "On one hand... on the other" situations
**Don't use when:**
- Cause-effect chain needed → Use Causal
- Explaining observation → Use Abductive
- Evaluating past decisions → Use Counterfactual
## Core Principles
### Charitable Interpretation
Each position must be represented at its strongest:
- Steel-man, don't straw-man
- Assume good faith and valid reasoning
- Identify the kernel of truth in each view
### Genuine Synthesis
Synthesis is NOT:
- Compromise (splitting the difference)
- Victory (one side wins)
- Avoidance (postpone decision)
Synthesis IS:
- Integration at higher level of abstraction
- Resolution that addresses underlying concerns
- New position that transcends original framing
## Three-Stage Process
### Stage 1: Thesis
**Purpose:** Articulate first position at maximum strength.
**Components:**
```yaml
thesis:
position:
statement: "Core claim being made"
underlying_concern: "What this position is really about"
stakeholder:
who: "Person/team holding this view"
role: "Their organizational function"
incentives: "What they optimize for"
evidence:
supporting:
- claim: "Evidence point"
source: "Where this comes from"
strength: 0.0-1.0
empirical: [DataPoint]
logical: [Argument]
implications:
if_adopted: "What happens if we go this way"
risks: [Risk]
benefits: [Benefit]
```
**Example:**
```yaml
thesis:
position:
statement: "We should prioritize enterprise features over SMB growth"
underlying_concern: "Revenue concentration and deal size efficiency"
stakeholder:
who: "Sales leadership"
role: "Revenue generation"
incentives: "ARR, deal size, quota attainment"
evidence:
supporting:
- claim: "Enterprise deals average $400K vs SMB $5K"
source: "Q3 sales data"
strength: 0.95
- claim: "Sales cost per $ revenue 5x lower for enterprise"
source: "CAC analysis"
strength: 0.85
empirical:
- "3 enterprise deals = entire SMB revenue"
- "Enterprise churn 3% vs SMB 8%"
implications:
if_adopted: "Focus engineering on enterprise features, reduce SMB investment"
risks:
- "Lose SMB market to competitors"
- "Revenue concentration risk"
benefits:
- "Higher margins"
- "Larger average deal"
```
### Stage 2: Antithesis
**Purpose:** Articulate counter-position at maximum strength.
**Process:**
1. Identify what thesis misses or undervalues
2. Find stakeholder with opposing view
3. Build strongest case for alternative
4. Identify where thesis assumptions break
**Components:**
```yaml
antithesis:
position:
statement: "Counter claim"
underlying_concern: "What this position is really about"
stakeholder:
who: "Person/team holding this view"
role: "Their organizational function"
incentives: "What they optimize for"
critique_of_thesis:
- assumption_challenged: "Thesis assumes X"
counter_evidence: "But actually Y"
- risk_identified: "Thesis ignores Z"
evidence:
supporting: [EvidencePoint]
empirical: [DataPoint]
logical: [Argument]
implications:
if_adopted: "What happens if we go this way"
risks: [Risk]
benefits: [Benefit]
```
**Example:**
```yaml
antithesis:
position:
statement: "SMB volume creates the foundation for sustainable growth"
underlying_concern: "Market presence, product iteration, and risk distribution"
stakeholder:
who: "Product leadership"
role: "Product-market fit and growth"
incentives: "Usage, retention, feature validation"
critique_of_thesis:
- assumption_challenged: "Enterprise features drive growth"
counter_evidence: "SMB usage generates product insights 10x faster"
- assumption_challenged: "Revenue concentration is acceptable"
counter_evidence: "Losing 1 enterprise deal = losing 80 SMB accounts"
- risk_identified: "Enterprise sales cycle is 9 months"
evidence:
supporting:
- claim: "SMB accounts generate 80% of feature requests"
source: "Product feedback analysis"
strength: 0.90
- claim: "SMB provides faster iteration cycles"
source: "Release metrics"
strength: 0.85
empirical:
- "SMB churn prediction accuracy 95% vs enterprise 60%"
- "Product improvements from SMB feedback shipped in 2 weeks"
implications:
if_adopted: "Maintain SMB investment, use as product lab"
risks:
- "Slower revenue growth short-term"
- "Lower margin overall"
benefits:
- "Diversified revenue base"
- "Faster product iteration"
- "Lower concentration risk"
```
### Stage 3: Synthesis
**Purpose:** Integrate positions at higher level, resolving underlying tensions.
**Synthesis Approaches:**
| Approach | When to Use | Example |
|----------|-------------|---------|
| **Integration** | Both positions address valid concerns | "Enterprise revenue + SMB as product lab" |
| **Sequencing** | Temporal resolution possible | "SMB first for PMF, then enterprise scale" |
| **Segmentation** | Different contexts warrant different approaches | "SMB for product X, Enterprise for product Y" |
| **Reframing** | Original dichotomy was false | "The real question isn't SMB vs Enterprise, it's time-to-value" |
| **Transcendence** | Higher goal subsumes both | "Optimize for sustainable unit economics regardless of segment" |
**Synthesis Components:**
```yaml
synthesis:
integrated_position:
statement: "What we will actually do"
framing: "How this resolves the tension"
how_thesis_is_addressed:
concern_validated: "What's true about thesis"
how_incorporated: "How we address that concern"
how_antithesis_is_addressed:
concern_validated: "What's true about antithesis"
how_incorporated: "How we address that concern"
trade_offs_acknowledged:
- trade_off: "What we're giving up"
mitigation: "How we reduce impact"
accepted_by: "Stakeholder who accepts this"
resolution_type: integration | sequencing | segmentation | reframing | transcendence
implementation:
actions: [Action]
metrics: [Metric] # How we know it's working
review_date: date # When we reassess
```
**Example:**
```yaml
synthesis:
integrated_position:
statement: "SMB as rapid learning engine, enterprise as revenue engine,
with explicit feature graduation path"
framing: "Not SMB vs Enterprise, but learning velocity vs revenue efficiency
with a bridge between them"
how_thesis_is_addressed:
concern_validated: "Enterprise deals are more efficient per dollar"
how_incorporated: "Maintain enterprise sales motion, prioritize enterprise
features that have been validated through SMB"
how_antithesis_is_addressed:
concern_validated: "SMB generates faster product learning"
how_incorporated: "Protect SMB investment as product lab, use SMB metrics
to prioritize enterprise features"
trade_offs_acknowledged:
- trade_off: "Some enterprise-only features will ship slower"
mitigation: "Identify 'must have' enterprise features, fast-track those"
accepted_by: "Sales leadership (with fast-track list)"
- trade_off: "Some SMB features won't graduate to enterprise"
mitigation: "Clear graduation criteria defined upfront"
accepted_by: "Product leadership (with criteria agreement)"
resolution_type: integration
implementation:
actions:
- "Define feature graduation criteria (Product + Sales)"
- "Create SMB → Enterprise feature pipeline"
- "Allocate 60% engineering to graduated features, 40% to SMB lab"
metrics:
- "SMB feature graduation rate (target: 3/month)"
- "Enterprise close rate on graduated features (target: +20%)"
- "Combined revenue growth (target: 30% QoQ)"
review_date: "End of Q2"
```
## Quality Gates
| Gate | Requirement | Failure Action |
|------|-------------|----------------|
| Thesis strength | Steel-manned, evidence-backed | Strengthen before proceeding |
| Antithesis genuine | Not straw-man, different stakeholder | Find genuine opposition |
| Synthesis integrative | Not compromise or victory | Reframe until true synthesis |
| Trade-offs explicit | All parties acknowledge costs | Surface hidden disagreements |
| Actionable | Concrete next steps | Add implementation detail |
## Stakeholder Agreement Protocol
Synthesis isn't complete until affected stakeholders acknowledge:
1. **Their concern was understood** (thesis/antithesis accurately represented)
2. **The synthesis addresses their core interest** (not just their stated position)
3. **They accept the trade-offs** (explicitly, not assumed)
```yaml
stakeholder_acknowledgment:
thesis_stakeholder:
name: "Sales leadership"
concern_understood: true
synthesis_addresses_concern: true
accepts_trade_offs: true
conditions: "Fast-track list for critical enterprise features"
antithesis_stakeholder:
name: "Product leadership"
concern_understood: true
synthesis_addresses_concern: true
accepts_trade_offs: true
conditions: "Clear graduation criteria before implementation"
```
## Common Failure Modes
| Failure | Symptom | Fix |
|---------|---------|-----|
| **False dichotomy** | Positions aren't truly opposed | Reframe the actual tension |
| **Straw-man** | Weak representation of one side | Involve actual stakeholder |
| **Mushy middle** | Synthesis is just "do both" | Force resource allocation |
| **Unacknowledged loss** | Trade-offs hidden | Surface what's being given up |
| **No implementation** | Synthesis is abstract | Add concrete actions |
## Output Contract
```yaml
dialectical_output:
thesis:
position: string
stakeholder: string
evidence: [EvidencePoint]
strength: float # 0.0-1.0
antithesis:
position: string
stakeholder: string
evidence: [EvidencePoint]
strength: float
synthesis:
position: string
resolution_type: string
confidence: float
integration:
thesis_addressed: string
antithesis_addressed: string
trade_offs:
- trade_off: string
mitigation: string
accepted_by: string
stakeholder_agreement:
- stakeholder: string
agrees: bool
conditions: optional<string>
implementation:
actions: [string]
metrics: [string]
review_date: date
next:
suggested_mode: ReasoningMode # Usually causal
canvas_updates: [string]
trace:
duration_ms: int
rounds_of_refinement: int
```
## Example Execution
**Context:** "Engineering wants to rebuild core platform (6 months). Sales wants new features for Q2 deals."
**Stage 1 - Thesis (Engineering):**
```
Position: "Technical debt is blocking velocity. Rebuild now or pay 10x later."
Evidence:
- Deploy time increased 300% YoY
- 40% of sprint spent on workarounds
- 3 critical bugs from architecture issues
Underlying concern: Sustainable development velocity
```
**Stage 2 - Antithesis (Sales):**
```
Position: "We have $2M in pipeline dependent on Q2 features. Delay = lose deals."
Evidence:
- 5 enterprise deals waiting on specific features
- Competitor launching similar features in March
- Q2 quota at risk without new capabilities
Underlying concern: Revenue target attainment
```
**Stage 3 - Synthesis:**
```
Integrated position: "Strangler fig pattern - rebuild incrementally while
delivering high-priority features"
How thesis addressed: Platform rebuild happens, but in modules alongside features
How antithesis addressed: Q2 features delivered, no delay
Trade-offs:
- Rebuild takes 9 months instead of 6 (Engineering accepts)
- Only top 3 features in Q2, not all 5 (Sales accepts with prioritization input)
Resolution type: Integration via sequencing
Implementation:
- Week 1: Joint prioritization session (top 3 features + first rebuild module)
- Q2: Deliver features on new modules where possible
- Q3-Q4: Complete rebuild with feature delivery continuing
```Related Skills
when-optimizing-agent-learning-use-reasoningbank-intelligence
Implement adaptive learning with ReasoningBank for pattern recognition, strategy optimization, and continuous improvement
reasoningbank-with-agentdb
Implement ReasoningBank adaptive learning with AgentDB's 150x faster vector database. Includes trajectory tracking, verdict judgment, memory distillation, and pattern recognition. Use when building self-learning agents, optimizing decision-making, or implementing experience replay systems.
reasoningbank-intelligence
Implement adaptive learning with ReasoningBank for pattern recognition, strategy optimization, and continuous improvement. Use when building self-learning agents, optimizing workflows, or implementing meta-cognitive systems.
reasoningbank-adaptive-learning-with-agentdb
Implement ReasoningBank adaptive learning with AgentDB for trajectory tracking, verdict judgment, memory distillation, and pattern recognition to build self-learning agents that improve decision-making through experience.
nowait-reasoning-optimizer
Implements the NOWAIT technique for efficient reasoning in R1-style LLMs. Use when optimizing inference of reasoning models (QwQ, DeepSeek-R1, Phi4-Reasoning, Qwen3, Kimi-VL, QvQ), reducing chain-of-thought token usage by 27-51% while preserving accuracy. Triggers on "optimize reasoning", "reduce thinking tokens", "efficient inference", "suppress reflection tokens", or when working with verbose CoT outputs.
reasoning-inductive
Extract patterns and generalizations from multiple observations. Use when detecting recurring themes, building predictive rules, or identifying systemic behaviors from accumulated data. Produces validated patterns with confidence bounds and exception handling.
reasoning-counterfactual
Evaluate alternative scenarios by simulating interventions on past decisions or hypothetical futures. Use when assessing decisions in hindsight, planning scenarios, or comparing paths not taken. Produces comparative analysis with probability-weighted outcomes.
reasoning-causal
Execute evidence-based decision-making through 6-stage causal flow. Use for known processes, operational execution, and decisions with clear cause-effect chains.
reasoning-analogical
Transfer knowledge from source domains to novel target situations through structural mapping. Use when facing new markets, products, or situations where past experience provides relevant patterns. Produces adapted solutions with explicit mappings and context adjustments.
reasoning-abductive
Generate and evaluate explanatory hypotheses from incomplete observations. Use when diagnosing anomalies, explaining unexpected outcomes, or inferring causes from effects. Produces ranked hypotheses with evidence and confidence scores.
meta-cognitive-reasoning
Meta-cognitive reasoning specialist for evidence-based analysis, hypothesis testing, and cognitive failure prevention. Use when conducting reviews, making assessments, debugging complex issues, or any task requiring rigorous analytical reasoning. Prevents premature conclusions, assumption-based errors, and pattern matching without verification.
azure-quotas
Check/manage Azure quotas and usage across providers. For deployment planning, capacity validation, region selection. WHEN: "check quotas", "service limits", "current usage", "request quota increase", "quota exceeded", "validate capacity", "regional availability", "provisioning limits", "vCPU limit", "how many vCPUs available in my subscription".