Overview
This skill enables Claude to simulate a “subconscious” execution layer that continuously analyzes, reflects, plans, and improves workflows in the background while actively completing tasks.
The workflow focuses on:
- hidden planning systems
- reflective reasoning
- iterative self-improvement
- context monitoring
- workflow optimization
- long-term execution refinement
- parallel internal analysis
- proactive problem detection
The goal is to improve:
- execution quality
- reasoning depth
- workflow consistency
- strategic planning
- long-term task coherence
Instead of reacting only to immediate prompts, Claude should continuously maintain deeper contextual awareness and improve workflows proactively.
Setup
Before starting:
- Define:
- primary workflow
- project goals
- active tasks
- optimization targets
- Create structured memory and reflection systems.
Recommended structure:
memory/
reflection/
optimization/
planning/
- Create files for:
- active context
- improvement opportunities
- recurring issues
- workflow summaries
- strategic observations
Recommended tools:
- Claude
- Markdown notes
- GitHub repositories
- Notion
- Obsidian
Optional:
- Vector memory systems
- Multi-agent orchestration
- Autonomous workflow schedulers
Inputs Required
- Active workflow or project
- Existing execution context
- Long-term goals
- Task history
- Current blockers
Optional:
- Session summaries
- Review feedback
- Memory systems
- Optimization logs
When to Use This Skill
Use this skill when:
- managing long-running projects
- coordinating complex workflows
- improving execution quality over time
- building autonomous agent systems
- maintaining strategic continuity
- optimizing repeated workflows
- scaling multi-step reasoning systems
When NOT to Use
Do NOT use this skill for:
- tiny one-step tasks
- purely reactive workflows
- disposable experiments
- tasks without long-term context
- highly constrained deterministic pipelines
Example Use Case
Build and maintain a large AI tooling repository over several months.
Claude should:
- Track project direction continuously
- Detect workflow inefficiencies
- Reflect on repeated mistakes
- Improve documentation structure
- Maintain long-term execution consistency
- Suggest proactive optimizations
- Refine workflows incrementally
Final result should:
- improve over time
- reduce repeated failures
- maintain project coherence
- strengthen planning quality
- support autonomous workflow refinement
Core Subconscious Principles
1. Maintain Background Reflection
Claude should continuously analyze:
- current execution quality
- workflow efficiency
- recurring issues
- strategic alignment
The system should:
- reflect while executing
- identify weaknesses proactively
- improve workflows incrementally
Good reflection improves:
- execution quality
- planning depth
- long-term consistency
2. Separate Execution From Reflection
Execution and reflection should operate as separate layers.
Example:
- foreground layer → active task execution
- subconscious layer → analysis and optimization
The subconscious layer should monitor:
- inefficiencies
- contradictions
- missed opportunities
- recurring patterns
This separation improves:
- reasoning depth
- workflow adaptability
- strategic awareness
3. Preserve Long-Term Context
Subconscious systems rely heavily on:
- memory continuity
- historical decisions
- workflow patterns
- accumulated knowledge
Claude should maintain:
- project summaries
- recurring lessons
- optimization history
- strategic direction
Long-term context improves:
- continuity
- planning quality
- workflow refinement
4. Improve Workflows Iteratively
The system should continuously ask:
- what can be simplified?
- what causes repeated problems?
- what should be automated?
- what slows execution?
Claude should:
- refine workflows gradually
- reduce friction over time
- optimize repeated systems
Iterative improvement compounds significantly over long projects.
5. Detect Problems Before Failure
Subconscious reasoning should proactively identify:
- weak plans
- inconsistent structure
- workflow bottlenecks
- scalability issues
- context fragmentation
Early detection improves:
- reliability
- maintainability
- execution stability
Workflow
1. Establish Active Context
Start by defining:
- current objectives
- project goals
- active workflows
- long-term direction
Create structured summaries for:
- ongoing tasks
- blockers
- recent decisions
- optimization targets
2. Create Reflection Systems
Maintain:
- workflow observations
- recurring mistakes
- improvement opportunities
- optimization ideas
Recommended files:
reflection/
workflow-observations.md
recurring-issues.md
optimization-log.md
The reflection layer should remain:
- lightweight
- structured
- continuously updated
3. Monitor Workflow Quality
Continuously analyze:
- execution speed
- planning quality
- repeated failures
- context consistency
- workflow complexity
Claude should identify:
- inefficient systems
- duplicated work
- unclear structure
- scaling problems
4. Generate Improvement Suggestions
The subconscious layer should suggest:
- simplifications
- automations
- structural improvements
- memory optimizations
- workflow refinements
Prioritize:
- high-impact improvements
- repeated pain points
- scalability gains
5. Maintain Strategic Continuity
Claude should continuously align work with:
- long-term goals
- project direction
- architectural consistency
- workflow philosophy
Avoid:
- fragmented execution
- inconsistent systems
- short-term optimization traps
6. Refine Systems Incrementally
Small improvements compound over time.
Claude should:
- improve documentation gradually
- simplify workflows continuously
- reduce repeated friction
- optimize execution patterns
Avoid:
- massive disruptive rewrites
- overengineering optimization systems
- unnecessary complexity
7. Review & Evolve
Regularly review:
- workflow effectiveness
- memory usefulness
- planning quality
- execution consistency
Refine:
- reflection systems
- optimization pipelines
- strategic workflows
- subconscious analysis patterns
The system should evolve continuously through iterative refinement.
Output Expectations
The final output should include:
- structured reflection systems
- optimization workflows
- long-term context continuity
- workflow refinement pipelines
- proactive planning systems
- execution quality improvements
The workflow itself should remain:
- lightweight
- scalable
- reflective
- adaptable
- continuously improving
Execution Strategy (for AI agents)
The agent should:
- Maintain separate reflection and execution layers
- Continuously analyze workflow quality
- Detect inefficiencies proactively
- Improve systems incrementally
- Preserve long-term context strategically
- Optimize execution consistency over time
The workflow should optimize for:
- reasoning depth
- long-term continuity
- execution quality
- workflow adaptability
- iterative self-improvement
Best Practices
- Keep reflection systems lightweight
- Improve workflows incrementally
- Preserve strategic continuity
- Detect repeated friction points early
- Maintain structured memory systems
- Separate execution from analysis
- Optimize for long-term consistency over short-term speed
Notes
- Reflective systems significantly improve long-term execution quality
- Small workflow improvements compound over time
- Good subconscious systems reduce repeated failures
- Long-term context continuity improves strategic reasoning
- The best optimization systems remain lightweight and adaptable