Coverage Workflow

The coverage workflow transforms a screenplay input through multiple analysis stages, culminating in a professional recommendation and comprehensive report. This guide provides a complete end-to-end walkthrough of the professional screenplay analysis process.

Contents

Workflow Overview
System Architecture
Scoring Flow
Decision Tree
Detailed Workflow
Workflow Summary

Workflow Overview

The coverage workflow transforms a screenplay input through multiple analysis stages, culminating in a professional recommendation and comprehensive report.

End-to-End Coverage Pipeline

Pipeline Stages

Input: Accept screenplay in multiple formats (FDX, PDF, plain text)
Parse: Extract scenes, dialogue, action lines, and character information
Analyze: Run comprehensive 16-point rubric evaluation using AI models
Score: Calculate weighted numerical scores and assign letter grades
Recommend: Determine final tier (PASS/CONSIDER/RECOMMEND) with rationale
Report: Generate professional coverage document in requested format

System Architecture

The coverage system is built on a layered architecture with clear separation between CLI interface, analysis engine, AI integration, and output formatting.

Coverage System Architecture

Architecture Layers

Layer	Responsibility	Key Components
CLI Layer	Command-line interface and user interaction	Cobra commands, flag parsing, input validation
Coverage Engine	Core analysis logic and coordination	Parser, evaluator, calculator, beat sheet generator
AI Integration	Language model communication and prompt management	vLLM client, prompt templates, response parsing
Output Formatters	Report generation in multiple formats	HTML renderer, PDF generator, JSON serializer, TUI display

Scoring Flow

Scores are calculated through a weighted aggregation system, combining individual rubric item scores into a final grade with recommendation tier.

Score-to-Grade Mapping

Score Range	Grade	Recommendation	Meaning
90-100	A	RECOMMEND	Excellent - Fast track for production
80-89	B	CONSIDER	Good - Strong consideration with enthusiasm
70-79	C	CONSIDER	Fair - Consider with reservations
60-69	D	PASS	Poor - Pass but track the writer
0-59	F	PASS	Not ready - Hard pass

Decision Tree

The final recommendation combines numerical scoring with qualitative red flag analysis to produce actionable guidance.

Decision Criteria

Score ≥ 90: Excellent screenplay. Check for critical red flags (legal, E&O issues). If clean, RECOMMEND (Grade A). If flags exist, downgrade to CONSIDER.
Score 70-89: Good to fair screenplay. CONSIDER with varying enthusiasm (Grade B or C) based on score and qualitative factors.
Score 60-69: Poor execution. Evaluate if concept is salvageable. If yes, PASS but track the writer (Grade D). If no, hard PASS (Grade F).
Score < 60: Not ready for production. Hard PASS (Grade F) unless exceptional circumstances exist.
Red Flags: Legal exposure, E&O concerns, rights issues, or controversial content can override score and force downgrade.

Detailed Step-by-Step Workflow

Follow this comprehensive workflow to execute professional screenplay coverage from initial input through final report generation.

Screenplay Acquisition

Download or provide the screenplay file in a supported format (FDX, PDF, or plain text).

# Download screenplay from public database
conductor screenplays download "Screenplay Title" --format fdx --output screenplay.fdx

# Or use existing file
cp /path/to/screenplay.fdx ./screenplay.fdx

Outputs: Screenplay file ready for parsing

Initial Parse & Validation

Extract screenplay structure including scenes, dialogue, action lines, and character information. Validate format compliance.

# Parse screenplay and validate format
conductor coverage format screenplay.fdx --compliance-check --detailed

# Expected output: Scene count, character list, page count, format issues

Outputs: Structured screenplay data, validation report, initial metadata

Beat Sheet Generation

Identify story structure using Save the Cat, Hero's Journey, or Three-Act methodology.

# Generate Save the Cat beat sheet
conductor coverage beatsheet screenplay.fdx --format save-the-cat --output beatsheet.json

# Generates: Opening Image, Setup, Catalyst, Debate, Break into Two, B Story, Fun and Games, Midpoint, Bad Guys Close In, All Is Lost, Dark Night, Break into Three, Finale, Final Image

Outputs: Beat sheet JSON with page numbers and analysis

16-Point Rubric Analysis

Execute comprehensive evaluation across all rubric categories using AI-powered analysis.

# Run full 16-point coverage analysis
conductor coverage full screenplay.fdx --detailed --output-format json --output coverage-data.json

# Or run individual rubric items:
conductor coverage logline screenplay.fdx
conductor coverage structure screenplay.fdx --analyze-acts --identify-beats
conductor coverage characters screenplay.fdx --analyze-arcs --dialogue-authenticity
conductor coverage themes screenplay.fdx --identify-primary --tone-consistency
# ... (all 16 items)

Outputs: Detailed analysis for each rubric item with scores and commentary

Score Calculation & Grade Assignment

Calculate weighted scores across all categories and assign letter grade.

# Calculate scores with detailed breakdown
conductor coverage score screenplay.fdx --detailed-breakdown --grade-assignment

# Output includes:
# - Individual category scores (0-100)
# - Category weights (Premise 20%, Characters 20%, Structure 20%, Dialogue 15%, Originality 10%, Commercial 15%)
# - Final weighted score
# - Letter grade (A, B, C, D, F)

Outputs: Numerical scores, weighted calculation, letter grade

Red Flag Analysis

Scan for legal exposure, E&O concerns, rights issues, and controversial content.

# Comprehensive risk analysis
conductor coverage risks screenplay.fdx --legal-scan --eo-analysis --detailed

# Checks for:
# - Defamation/libel risks
# - Real person/entity depictions
# - Intellectual property concerns
# - Controversial subject matter
# - E&O insurance considerations

Outputs: Red flag report with severity levels and mitigation recommendations

Recommendation Tier Determination

Combine score, grade, and red flag analysis to produce final recommendation tier.

# Generate recommendation with rationale
conductor coverage recommend screenplay.fdx --calculate-tier --generate-rationale --detailed

# Decision logic:
# Score 90-100 + No Critical Flags = RECOMMEND (Grade A)
# Score 80-89 OR High Score with Flags = CONSIDER (Grade B)
# Score 70-79 = CONSIDER with reservations (Grade C)
# Score 60-69 = PASS, track writer (Grade D)
# Score < 60 = PASS, hard pass (Grade F)

Outputs: Recommendation tier (PASS/CONSIDER/RECOMMEND), rationale, next steps

Action Plan Generation

Create concrete next steps with timeline and resource estimates.

# Generate actionable next steps
conductor coverage action screenplay.fdx --generate-steps --timeline-estimate --resource-allocation

# For RECOMMEND: Fast-track process, option negotiation, talent attachment
# For CONSIDER: Development meetings, writer notes, second read
# For PASS: Polite decline, writer tracking (if applicable)

Outputs: Action items, timeline, required resources, responsible parties

Report Generation

Compile all analysis into professional coverage report in requested format.

# Generate HTML report
conductor coverage full screenplay.fdx --output-format html --output coverage-report.html

# Generate PDF report
conductor coverage full screenplay.fdx --output-format pdf --output coverage-report.pdf

# Generate JSON data
conductor coverage full screenplay.fdx --output-format json --output coverage-data.json

# Display in terminal
conductor coverage full screenplay.fdx --output-format terminal

Outputs: Professional coverage report with all rubric items, scores, and recommendations

Quality Assurance & Benchmarking

Verify analysis quality and benchmark performance against industry standards.

# Run quality benchmarks
conductor coverage benchmark screenplay.fdx --iterations 5 --warmup 2 --memory --quality-gates

# Validates:
# - Analysis accuracy (comparison to ground truth if available)
# - Performance metrics (latency, throughput)
# - Memory usage
# - Quality gate compliance (>= 90% accuracy, < 2s latency)

Outputs: Benchmark results, quality gate pass/fail, performance metrics

Progress Tracking (Optional)

Compare coverage across multiple drafts to track improvement.

# Track progress against baseline
conductor coverage track screenplay-v2.fdx --compare coverage-v1.json

# Shows delta in:
# - Individual rubric scores
# - Overall grade
# - Recommendation tier changes
# - Specific improvements/regressions

Outputs: Comparison report showing changes between drafts

Workflow Summary

Key Takeaways

Coverage workflow follows a systematic 11-step process from acquisition to final report
Each step produces specific outputs that feed into subsequent stages
Scoring combines numerical analysis (0-100) with qualitative assessment (red flags)
Decision tree produces actionable recommendations (PASS/CONSIDER/RECOMMEND)
Architecture separates concerns: CLI → Engine → AI → Output formatters
Quality assurance validates accuracy and performance against industry standards
Workflow supports both one-time analysis and multi-draft tracking

Best Practices

Always validate format compliance before running full analysis
Generate beat sheet early to inform structural analysis
Review red flags manually before finalizing recommendation
Use JSON output format for programmatic integration
Run benchmarks periodically to ensure consistent quality
Track progress across drafts for revision guidance
Combine automated analysis with human expertise for final decisions