On Practice Methodology Toolkit

Practical methodology toolkit derived from Mao Zedong's On Practice. Transforms the philosophical theory of the unity of knowing and doing into actionable, reusable operational frameworks for decision-making, problem analysis, and report writing.

Core Principle: Practice is the source, driving force, purpose, and sole criterion for testing truth. Cognition follows a spiral: practice → perceptual knowledge → rational knowledge → re-practice → re-cognition, repeating infinitely. All correct ideas come from social practice; their real value lies in guiding and improving practice.

Prerequisites

• No runtime dependencies (methodology-only skill, no scripts)
• Works in any domain — business, engineering, personal, strategic

When to Use

Invoke when:

• User needs to make a decision grounded in real-world evidence rather than speculation
• User faces a problem that requires investigation and fact-finding before analysis
• User wants to write a report structured around what was done, learned, and will be improved
• User mentions "practice-based", "verify through practice", "seek truth from facts", "test assumptions"
• User needs to move from theory to action, or validate ideas through experimentation
• User wants to break analysis paralysis by grounding thinking in concrete practice

Do NOT invoke when:

• User needs contradiction-based structural analysis → use on-contradiction
• User needs code implementation → use requirement-workflow or IDE directly
• User needs software-specific methodology → use software-methodology-toolkit
• The problem is purely theoretical with no practical component

Relationship with On Contradiction and On Protracted War

On Practice, On Contradiction, and On Protracted War form a trilogy:

Dimension	On Contradiction	On Practice	On Protracted War
Focus	Structure of forces	Process of knowing	Evolution over time
Question	"What are the contradictions?"	"How do we verify truth?"	"How does this unfold and when do we act?"
Method	Identify, prioritize, transform	Investigate, test, validate	Stage, strategize, maneuver, evolve
Strength	Structural clarity — sees the skeleton	Process rigor — ensures grounding	Temporal wisdom — sees the arc of change
Combine	Identify WHAT forces are at play	Validate HOW through practice	Plan WHEN to act and how each phase differs

The Practice-Cognition Spiral

Every agent in this toolkit applies this spiral:

Step 1: PRACTICE    — Engage with reality: investigate, experiment, observe
Step 2: PERCEIVE    — Gather perceptual knowledge: raw facts, data, impressions
Step 3: REASON      — Elevate to rational knowledge: patterns, laws, theories
Step 4: RE-PRACTICE — Apply rational knowledge back to practice: test, validate
Step 5: RE-COGNIZE  — Refine understanding based on results: correct, deepen
                      ↻ Repeat — each cycle spirals upward

This spiral is the DNA of all agents. Each applies it to a specific domain.

Key Concepts

Concept	Definition	Practical Meaning
Practice is Primary	All knowledge originates from practice	Don't theorize in a vacuum; get your hands dirty first
Perceptual Knowledge	Direct, surface-level impressions from experience	Raw data, observations, first-hand impressions — necessary but insufficient
Rational Knowledge	Systematic understanding of patterns and laws	Theories, frameworks, principles extracted from perceptual knowledge
Perceptual → Rational Leap	The qualitative leap from observation to understanding	Don't stay at the data level; synthesize patterns and laws
Rational → Practice Return	The second leap: applying theory back to action	Theory without practice is empty; apply and validate
Unity of Knowing and Doing	Knowledge and action are inseparable	Knowing without doing is not truly knowing
Seek Truth from Facts	Conclusions must come from actual investigation	No investigation, no right to speak
Spiral Development	Each practice-cognition cycle advances understanding	Learning is not linear; it spirals upward through repetition
Practice as Criterion	Only practice can verify whether an idea is correct	Arguments don't settle truth — results do
Concrete Analysis	Study the actual conditions, not abstract categories	Every situation has specific circumstances that generic theory misses

Agent Summary

Domain	Agent	Core Principles Applied
Thinking	decision-maker	Small-scale trials + feedback validation + unity of judgment and results
Thinking	problem-analyzer	Investigation first + perceptual → rational leap + internal process tracing
Writing	report-writer	Practice-based evidence chain + continuous improvement cycle

Routing Decision Table

User Signal	Agent	Confidence
"make a decision", "choose between", "test this option", "validate assumptions"	decision-maker	High
"analyze the problem", "investigate", "what actually happened", "root cause"	problem-analyzer	High
"write a report", "summarize findings", "document what we learned"	report-writer	High
"实事求是", "调查研究", "实践检验"	problem-analyzer	High
Need to validate an idea before committing	decision-maker	Medium
Need to present practice-based findings to stakeholders	report-writer	Medium
General mention of "practice" or "On Practice"	problem-analyzer (default entry)	Low

If confidence is Low, confirm agent selection with the user before proceeding.

Composition Workflows

Full Practice Cycle Workflow (Investigate → Decide → Report)

1. problem-analyzer  → Investigate the real situation, extract rational knowledge
2. decision-maker    → Design practice-based validation, decide based on evidence
3. report-writer     → Document what was done, learned, verified, and next steps

Assumption Validation Workflow

1. decision-maker    → Identify assumptions, design small-scale tests
2. problem-analyzer  → Analyze test results, extract patterns
3. decision-maker    → Refine decision based on validated evidence

Continuous Improvement Workflow

1. problem-analyzer  → Investigate current state through practice lens
2. report-writer     → Document the practice → learning → improvement cycle
   ↻ Repeat each iteration

Cross-Skill Composition Workflows (Trilogy)

Validate-Then-Structure (with On Contradiction)

1. on-practice / problem-analyzer       → Investigate reality: gather first-hand perceptual knowledge
2. on-practice / decision-maker         → Audit assumptions: which have practice evidence?
3. on-contradiction / problem-analyzer  → Structure validated findings into contradiction layers
4. on-contradiction / decision-maker    → Identify principal contradiction, plan transformation
5. on-practice / report-writer          → Document the full practice → structure → action cycle

Practice-Driven Decision (with On Contradiction)

1. on-practice / decision-maker         → Surface assumptions, design small-scale trials
2. on-contradiction / decision-maker    → Frame validated results as contradictions, find principal one
3. on-practice / decision-maker         → Final decision grounded in both evidence and structural clarity

The Full Trilogy Workflow (Evidence → Structure → Time)

1. on-practice / problem-analyzer        → Investigate reality, gather first-hand perceptual knowledge
2. on-contradiction / problem-analyzer   → Structure findings into contradiction layers, find principal one
3. on-protracted-war / problem-analyzer  → Diagnose current stage, assess four factors over time
4. on-protracted-war / decision-maker    → Choose phase-appropriate strategy
5. on-practice / report-writer           → Full report: evidence + contradiction structure + phased plan

Practice Analysis Tools

Tool 1: Practice-Cognition Map

For any situation, trace the knowledge chain:

Stage	Content	Evidence	Confidence
Practice (what was done)	{concrete actions taken}	{records, logs, data}	Factual
Perception (what was observed)	{raw observations}	{direct experience}	Perceptual
Reason (what was understood)	{patterns, principles extracted}	{analysis, synthesis}	Rational
Validation (was it correct?)	{re-practice results}	{outcomes, metrics}	Verified / Unverified

Tool 2: Assumption Audit

Before any decision, list all assumptions and their practice-basis:

#	Assumption	Practice-Based?	Evidence	Confidence
1	{assumption}	✅ Tested / ❌ Untested	{what evidence exists}	High / Low
2	{assumption}	✅ / ❌	{evidence}	High / Low

Rule: Untested assumptions with high impact must be validated through practice before proceeding.

Tool 3: Investigation Checklist

Before drawing ANY conclusion, verify:

1. Have you investigated the actual situation first-hand?
2. Is your data from practice (observed/measured) or from theory (assumed/deduced)?
3. Have you distinguished perceptual knowledge (raw facts) from rational knowledge (interpreted patterns)?
4. Has your conclusion been tested in practice, or is it still theoretical?
5. If tested, did the practice results confirm or contradict your theory?

"No investigation, no right to speak." — Mao Zedong

Agent Output Contract

All agents follow the same output rules:

Allowed	Not Allowed
Practice-grounded analysis with evidence	Conclusions without investigation evidence
Recommendations with validation plans	Pure theoretical reasoning without practice links
Experiment designs for testing assumptions	Presenting untested assumptions as facts
Actionable next practice steps	Abstract advice disconnected from action

Every agent output must include:

1. Practice Evidence — What was actually done/observed (not assumed)
2. Knowledge Stage — Whether findings are perceptual, rational, or practice-verified
3. Assumptions Audit — Which conclusions are tested vs untested
4. Validation Plan — How to test unverified conclusions through practice
5. Next Practice — Concrete actions for the next spiral cycle

Error Handling

Issue	Solution
User's request matches no agent trigger	Default to problem-analyzer as the entry point
User wants to decide without any evidence	Highlight untested assumptions; recommend small-scale practice first
Analysis is based entirely on theory/logic	Challenge: "What practice evidence supports this?" Design investigation
User cannot access real-world data	Suggest proxy practices: interviews, prototypes, small experiments
Previous practice results contradict current theory	This is the spiral at work — revise rational knowledge, plan re-practice
User expects certainty from limited practice	Clarify: practice spirals upward; one cycle gives partial truth, not full truth
Conclusions feel correct but are untested	Mark as "rational knowledge (unverified)" — design validation practice

Execution Checklist

Before invoking any agent, verify:

• The situation involves real-world action, not pure abstract reasoning
• Agent selection follows the Routing Decision Table
• Agent receives concrete context (what was done, what was observed, what is planned)

After agent produces output, verify:

• Every conclusion traces back to practice evidence (not just logic)
• Assumptions are explicitly labeled as tested or untested
• A validation plan exists for untested conclusions
• Next steps are concrete actions (practice), not abstract recommendations
• The output identifies which stage of the spiral the user is in

Boundary Enforcement

This skill ONLY handles:

• Practice-grounded decision-making with evidence validation
• Problem investigation rooted in real-world observation and data
• Report writing structured around the practice-cognition-improvement cycle
• Designing experiments and investigations to test assumptions
• Tracing the practice → cognition → re-practice spiral

This skill does NOT handle:

• Structural contradiction analysis → on-contradiction
• Long-term phased strategy or stage diagnosis → on-protracted-war
• Code generation or software implementation → requirement-workflow
• Software engineering methodology → software-methodology-toolkit
• Quick factual answers → answer directly
• Project management execution → out of scope (future v2.0)

Expansion Roadmap (v2.0)

Guiding principle: No new agents until v1.0 agents are practice-tested. Apply our own spiral — use the skills in real work first, then expand based on evidence.

The following agents are planned for future versions. Agents are assigned to the skill whose core method (practice spiral vs contradiction structure) is the primary lens.

Domain	Agent	Core Principle	Notes
Learning	learning-master	Learn by doing; deepen through repeated practice; summarize laws	Complements on-contradiction / knowledge-absorber (which focuses on structural universality/particularity)
Experimentation	experiment-designer	Small practices to test ideas; iterate on results
Execution	execution-driver	Translate plans into actions; correct deviations through practice
Investigation	field-investigator	First-hand research; rational knowledge from perceptual facts
Improvement	pattern-extractor	Extract regular patterns; turn experience into methods into systems	Complements on-contradiction / retrospective-guide (which focuses on contradiction development over time)

References

• On Practice (《实践论》) — Mao Zedong (1937)
• On Contradiction (《矛盾论》) — Mao Zedong (1937)
• On Protracted War (《论持久战》) — Mao Zedong (1938)
• "Where Do Correct Ideas Come From?" (《人的正确思想是从哪里来的？》) — Mao Zedong (1963)
• "Oppose Book Worship" (《反对本本主义》) — Mao Zedong (1930)