AgentSkillsCN

protocol-comprehensive

针对需要档案级高质量记录的重大作品,采用完整的分析处理流程(P7)。可通过“全面分析”、“综合拆解”、“一切尽在其中”、“完整覆盖”、“教学案例研究”、“档案级记录”等短语触发。产出全部 8 份核心文档及补充材料。激活所有 9 位分析师角色。预计每 100 页剧本耗时 8–12 小时。这正是完整的剧本分析与戏剧性结构分析技能的全面启动。

SKILL.md
--- frontmatter
name: protocol-comprehensive
description: Full analytical treatment protocol (P7) for major works requiring archival-quality documentation. Triggers on "full analysis", "comprehensive breakdown", "everything", "complete coverage", "teaching case study", "archival documentation". Produces all 8 core documents plus supplements. Activates all 9 analyst roles. Estimated time 8-12 hours per 100-page script. This is the complete script-analysis-dramaturgical skill activation.

P7 — COMPREHENSIVE Protocol

Full analytical treatment for major work or archival purposes.

Configuration

yaml
ROLES_ACTIVE:
  PRIMARY:   [ALL NINE ROLES]
  SECONDARY: []
  INACTIVE:  []

DOCUMENTS:
  GENERATE:  [ALL EIGHT CORE DOCUMENTS + SUPPLEMENTS AS WARRANTED]
  EXCLUDE:   []

DEPTH: Maximum
INGESTION: Four-phase protocol

This Protocol = Full SKILL.md Activation

P7 activates the complete script-analysis-dramaturgical skill. See /mnt/project/SKILL.md for full specification.

Analyst Role Matrix

All nine perspectives active:

RoleFocusMinimum Observations
AESTHETEForm, beauty, style3
DRAMATURGISTStructure, rhythm, tension5
NARRATOLOGISTMechanism, theory applicationFramework complete
ART_HISTHistorical, intertextual contextAs relevant
CINEPHILEComparable works3
RHETORICIANLanguage, argument, dialogueAssessment complete
PRODUCERPractical considerationsIf requested
ACADEMICTheoretical rigorMaintained throughout
FIRST-READEREngagement responseHonest assessment

Four-Phase Ingestion

code
PHASE 1: INITIAL READ
├── Read script completely without annotation
├── Note immediate emotional/engagement responses
└── Capture [FIRST-READER] reactions

PHASE 2: STRUCTURAL READ  
├── Identify act breaks (even if unconventional)
├── Mark turning points and structural nodes
└── Note temporal structure and time span

PHASE 3: ANALYTICAL READ
├── Annotate scene-by-scene
├── Tag every beat with function
└── Map character arcs and transformations

PHASE 4: SYNTHESIS
├── Cross-reference observations
├── Identify patterns across readings
└── Formulate diagnostic assessments

Eight Core Documents

#DocumentPurpose
1Coverage ReportExecutive summary + recommendation
2Beat MapExhaustive scene-by-scene mapping
3Structural AnalysisAct/movement architecture
4Character AtlasAll characters + arcs
5Thematic ArchitectureTheme layers + motif tracking
6Diagnostic ReportStructural problems + solutions
7Theoretical CorrespondenceFramework mapping
8Revision RoadmapPrioritized action items

Optional Supplements

Generate as warranted by work:

SupplementWhen to Include
Production NotesBudget/casting/feasibility relevant
Comparative AnalysisGenre positioning important
Dialogue AnalysisLanguage patterns notable
Visual GrammarCamera/staging/spectacle significant
Market PositioningCommercial assessment requested

Framework Application (Full)

Apply all established algorithms:

AlgorithmTests
McKeeGap analysis, progressive complications, controlling idea
AristotleUnity, catharsis, recognition, reversal
South ParkBut/Therefore causal binding
Phoebe Waller-BridgeThree-obstacle layering per scene
Larry DavidCascading consequences
Bharata MuniRasa identification, bhava mapping
KubrickNon-submersible units present?
Attention MechanicsMode consistency, engagement diagnosis

Quality Checklist

Before delivery:

code
â–¡ Every scene/beat is mapped (no gaps)
â–¡ All named characters are profiled
â–¡ All turning points are identified
â–¡ Diagnostic questions are answered with evidence
â–¡ Issues include specific line/page references
â–¡ Revision items include actionable suggestions
â–¡ ASCII diagrams render correctly
â–¡ All nine roles have contributed observations
â–¡ All relevant frameworks applied
â–¡ Cross-references to comparable works included

Template References

  • /mnt/project/coverage-template.md
  • /mnt/project/beat-map-template.md
  • /mnt/project/structural-template.md
  • /mnt/project/character-template.md
  • /mnt/project/diagnostic-template.md

Delivery Format

All documents delivered as markdown with:

  • ASCII diagrams for visual structures
  • Tables for data-dense sections
  • Code blocks for formulas and tests
  • Consistent heading hierarchy
  • Role-tagged observations throughout