non-fiction-revision
npx skills add https://github.com/jwynia/agent-skills --skill non-fiction-revision
Agent 安装分布
Skill 文档
Non-Fiction Revision Diagnostic
Purpose
Diagnose and guide revisions in non-fiction books (business, self-help, academic, popular science, memoir). Non-fiction operates across multiple levels simultaneouslyâthesis, structure, evidence, pedagogy. Changes at one level cascade to others. This skill identifies which level needs attention and prevents revision damage.
Quick Reference
| State | Signal | Core Issue |
|---|---|---|
| NR1 | Thesis feels unclear or weak | Conceptual level problem |
| NR2 | Arguments don’t build logically | Structural problem |
| NR3 | Claims lack adequate support | Evidence problem |
| NR4 | Readers report confusion | Pedagogical problem |
| NR5 | Sources outdated or weak | Credibility problem |
| NR6 | Changes causing new problems | Cascade failure |
The Multi-Level Structure
Every non-fiction book operates simultaneously across:
Level 1: Conceptual/Thesis
- Core thesis and main arguments
- Foundational assumptions and frameworks
- Philosophical or theoretical positions
- Overall purpose and intended impact
Level 2: Structural/Argument
- Logical argument sequence and flow
- Chapter organization and dependencies
- Evidence distribution and support patterns
- Reader journey and persuasion architecture
Level 3: Content/Evidence
- Specific evidence, examples, and data
- Explanations and clarifications
- Voice, tone, and accessibility
- Citations, sources, and credibility markers
Level 4: Pedagogical/Reader
- Learning progression and scaffolding
- Cognitive load management
- Engagement and retention strategies
- Practical application guidance
Critical Principle
Any change at one level potentially affects argument validity and reader comprehension at all other levels.
Changes propagate:
- Upward: New evidence might undermine existing arguments
- Downward: Thesis changes require complete restructuring
- Lateral: Chapter reordering affects argument development flow
Diagnostic States
NR1: Thesis/Core Argument Problem
Symptoms:
- Thesis feels unclear or unstated
- Main arguments seem weak or unconvincing
- Central claims are frequently disputed
- Book’s purpose is hard to articulate
- Conclusion doesn’t deliver on promise
Diagnostic Questions:
- Can you state the thesis in one sentence?
- What are the 3-5 main arguments supporting it?
- What evidence backs each main argument?
- Does the conclusion match the opening promise?
- What changes if the thesis is wrong?
Interventions:
- Map all current evidence supporting the thesis
- Identify which chapters depend on current thesis formulation
- Evaluate what new evidence a revised thesis would require
- Check how thesis change affects the reader promise
- Assess impact on conclusion and call-to-action
Cascade Warning: Thesis changes are the most dangerousâthey can invalidate entire chapters. Before changing thesis, map all dependencies.
NR2: Structural/Organization Problem
Symptoms:
- Argument flow feels illogical
- Chapters seem disconnected
- Prerequisite knowledge isn’t established before use
- Reader must jump back to understand
- Same points repeated without building
Diagnostic Questions:
- What logical prerequisites exist for each chapter?
- Does knowledge accumulate or repeat?
- Are there arguments that depend on later chapters?
- Could chapters be reordered without breaking logic?
- Do transitions explain the logical connection?
Interventions:
- Create dependency map of chapter relationships
- Identify logical prerequisites for each major argument
- Map reader knowledge accumulation through structure
- Test alternative sequences against comprehension requirements
- Ensure examples remain contextually appropriate
Cascade Warning: Reordering chapters affects every cross-reference and forward/backward reference. Track all internal citations.
NR3: Evidence/Support Problem
Symptoms:
- Claims feel unsupported or handwavy
- “Trust me” rather than “here’s proof”
- Evidence exists but doesn’t connect to claims
- Support for different claims inconsistent in quality
- Key arguments rest on weak foundations
Diagnostic Questions:
- What type of evidence supports each main claim?
- Are evidence standards consistent throughout?
- Which claims have the weakest support?
- Does evidence actually prove what’s claimed?
- Are there counter-arguments addressed?
Interventions:
- Audit evidence quality for each major claim
- Identify claims needing stronger or different support
- Evaluate how new evidence affects existing arguments
- Check consistency in citation style and source quality
- Assess whether adding evidence changes argument strength
Cascade Warning: Adding strong evidence for one claim can accidentally weaken others by raising the evidence standard readers expect.
NR4: Pedagogical/Comprehension Problem
Symptoms:
- Readers report confusion
- Complex concepts introduced too fast
- Examples don’t illuminateâthey confuse
- Practical application unclear
- Target audience can’t follow
Diagnostic Questions:
- Where do readers typically get lost?
- Are technical terms defined before use?
- Do examples match reader experience?
- Is cognitive load distributed or front-loaded?
- Can readers apply what they learn?
Interventions:
- Map cognitive load distribution across chapters
- Identify concepts needing better scaffolding
- Evaluate example effectiveness and relevance
- Check that practical guidance is actionable
- Assess whether complexity progression is appropriate
Cascade Warning: Simplifying can accidentally remove nuance that supports arguments. Balance accessibility with accuracy.
NR5: Credibility/Source Problem
Symptoms:
- Sources feel outdated
- Citation patterns inconsistent
- Author authority questioned
- Examples from wrong era
- “According to experts” without naming them
Diagnostic Questions:
- How old are the oldest sources?
- Are sources appropriate to the field?
- Is citation style consistent throughout?
- Do you name experts or speak generally?
- Are there sources readers would expect that are missing?
Interventions:
- Audit source currency and quality
- Identify claims needing more recent support
- Standardize citation style and depth
- Replace generic expert references with specific citations
- Add expected canonical sources for the field
Cascade Warning: Updating sources can accidentally change what the evidence actually says. Verify new sources support the same conclusions.
NR6: Cascade Failure
Symptoms:
- Fixes create new problems
- Changes in one chapter break another
- Evidence update invalidates argument
- Structural change creates new confusion
- Progress feels impossible
Diagnostic Questions:
- What was the original change that started the cascade?
- What dependencies weren’t mapped?
- Is there a stable rollback point?
- What’s the minimum viable change that tests the idea?
- Are the problems localized or systemic?
Interventions:
- Stop implementing and assess damage scope
- Identify last stable state
- Map actual dependencies (not assumed ones)
- Consider whether original change is worth the cascade cost
- If proceeding, create checkpoint system for controlled changes
Rollback Criteria:
- Fundamental logical structure breaks down
- Evidence requirements become impossible to meet
- Changes create more credibility problems than they solve
- Reader comprehension significantly compromised
Pre-Change Protocol
Before implementing ANY revision:
1. Identify Change Level
- Conceptual (thesis, main arguments, frameworks)
- Structural (chapter sequence, argument flow, organization)
- Content (evidence, examples, explanations)
- Pedagogical (presentation, scaffolding, application)
2. Map Dependencies
For each change, document:
- Prerequisites: What must remain intact for this change to work?
- Dependents: What later claims rely on this element?
- Evidence: What support becomes necessary or obsolete?
- Comprehension: How does this affect the learning journey?
3. Assess Cascade Risk
- Low: Change is isolated, no dependencies
- Medium: 2-3 other elements need updating
- High: Affects multiple chapters or core arguments
- Critical: Threatens book’s foundational structure
4. Define Success Criteria
Before changing, know how you’ll evaluate:
- Logical coherence: Does the argument still flow?
- Evidence adequacy: Are claims still supported?
- Reader comprehension: Can they still follow?
- Credibility: Does authority remain intact?
Change Record Template
# Revision: [Brief Description]
## Change Type
- [ ] Conceptual - [ ] Structural - [ ] Content - [ ] Pedagogical
## Rationale
[Why this change improves the book]
## Dependency Analysis
- Prerequisites affected:
- Dependent elements:
- Evidence changes needed:
- Comprehension impacts:
## Cascade Risk Level
- [ ] Low - [ ] Medium - [ ] High - [ ] Critical
## Success Criteria
- Logical coherence check:
- Evidence adequacy standard:
- Reader comprehension benchmark:
## Implementation Status
- [ ] Initial change complete
- [ ] Dependencies updated
- [ ] Cross-references revised
- [ ] Cascade effects resolved
## Outcome
[Complete after implementation]
Non-Fiction Type Variations
Academic/Research
- Methodology consistency paramount
- Literature review must stay current
- Anticipate peer review critique
- Contribution clarity essential
Business/Self-Help
- Practical applicability above all
- Examples must feel current
- Implementation guidance required
- ROI/benefit must be clear
Popular Science
- Accessibility without dumbing down
- Research currency matters
- Analogies must actually illuminate
- Balance education with engagement
Memoir/Personal Narrative
- Factual accuracy + compelling narrative
- Emotional authenticity preserved
- Privacy boundaries respected
- Personal connects to universal
Anti-Patterns
The Endless Revision Spiral
Fixing one thing breaks another, which breaks another. The book never reaches stable state.
Fix: Define minimum viable change, implement, stabilize before next change.
The Evidence Addiction
Adding more and more sources without improving argument quality. Quantity masking weakness.
Fix: Better evidence, not more evidence. One strong study beats ten weak ones.
The Clarity Trap
Simplifying until accuracy suffers. Readers can follow but learn the wrong thing.
Fix: Scaffold complexity rather than remove it. Build up to nuance.
The Thesis Drift
Small changes accumulate until the book argues something different than intended.
Fix: Regularly check: does the conclusion still match the introduction’s promise?
Integration Points
Inbound:
- From
research: When gathering new evidence - From
revision: For overall revision strategy
Outbound:
- To
prose-style: After structural issues resolved - To
fact-check: For evidence verification
Complementary:
research: For evidence gatheringrevision: For fiction revision (parallel skill)