pf-policy-comparison

Important: All documents generated by this plugin are drafts intended as starting points. Insurance documents, analyses, and recommendations must be reviewed by a licensed insurance professional before use with clients or carriers. This plugin does not provide insurance advice.

Side-by-side insurance policy comparison, gap analysis, and client recommendation letter generation for independent insurance agents.

An alternative to EZLynx and HawkSoft — a fully local, AI-powered workflow. No vendor lock-in. No subscription.

---

Installation

1. Download the pf-policy-comparison.plugin file
2. Open Claude Desktop and navigate to Settings > Plugins
3. Click Install Plugin and select the downloaded .plugin file
4. The plugin will be installed and available immediately

Note: All data stays local on your machine. No external API calls or cloud storage required.

Quick Start

1. Run /pc:pc-setup — initialize workspace folders
2. Copy policy declaration files (.txt, .pdf, or .docx) to inbound/
3. Run /pc:pc-run — full pipeline in one command

Or run stages individually:

/pc:pc-ingest       → extract coverage data from declaration files
/pc:pc-compare      → generate side-by-side comparison matrix (DOCX)
/pc:pc-gap-analysis → identify coverage gaps vs. industry benchmarks
/pc:pc-recommend    → draft client recommendation letter (DOCX + .txt)
/pc:pc-status       → check pipeline progress at any time

---

Commands

Command	Description
/pc:pc-setup	Initialize workspace — run once after installation
/pc:pc-ingest	Extract coverage data from policy declaration files in inbound/
/pc:pc-compare	Generate side-by-side DOCX comparison matrix
/pc:pc-gap-analysis	Identify gaps vs. homeowners/auto/commercial BOP benchmarks
/pc:pc-recommend	Draft personalized client recommendation letter (DOCX + .txt)
/pc:pc-run	Run the full pipeline end-to-end (supports --resume)
/pc:pc-status	Show stage progress and output file inventory
/pc:pc-kb	Display insurance domain knowledge base reference

---

Workflow Lifecycle Folders

your-workspace/
├── inbound/          ← Put policy declaration files here (.txt, .pdf, .docx)
├── processing/       ← Intermediate JSON data files (auto-generated)
├── outbound/         ← Final deliverables: comparison matrix + recommendation letter
├── archive/          ← Move old runs here to keep workspace clean
└── .pc/              ← Plugin config, stage checkpoints, logs

---

Supported Policy Types

• Homeowners (Coverage A/B/C/D/E/F)
• Auto (BI/PD/Collision/Comprehensive/UM-UIM)
• Renters
• Commercial BOP (General Liability, BPP, Building)
• Umbrella
• Other (best-effort extraction)

Carriers tested: State Farm, Allstate, Progressive, Travelers, Farmers, Liberty Mutual, USAA, Nationwide, Erie. Any carrier's declarations page works — coverage terminology is normalized automatically.

---

Feature Comparison

Feature	pf-policy-comparison	EZLynx	HawkSoft
Side-by-side comparison matrix	✅	✅	✅
Gap analysis vs. benchmarks	✅	✅	LiProprietaryed
Client recommendation letter	✅	LiProprietaryed	✅
Offline / local processing	✅	❌	❌
No subscription fee	✅	❌	❌
DOCX output	✅	✅	✅
Plain-text email copy	✅	❌	LiProprietaryed
Max policies per run	8	UnliProprietaryed	UnliProprietaryed
Multi-line policy types	✅	✅	✅
ISO form-aware analysis	✅	❌	❌
E&O-safe recommendation language	✅	❌	❌
Source-cited benchmarks (III, ISO, NAIC)	✅	❌	❌
Exclusion gap analysis with endorsement recs	✅	❌	❌
State-specific auto liability comparison	✅	✅	❌
Coinsurance penalty detection (HO-3)	✅	❌	❌

---

AI-Powered Features

• Coverage Gap Identification: Analyzes policies against III-sourced industry benchmarks to identify underinsured categories and missing coverages
• Risk Severity Ranking: Flags gaps by severity (critical, moderate, minor) with plain-language risk explanations for agents
• Side-by-Side Comparison Matrix: Generates professional DOCX comparison tables with ISO form-aware headers and coverage completeness indicators
• Benchmark-Based Analysis: Applies III-recommended coverage minimums with source citations (ISO HO-3 Section I Condition E coinsurance, IRC 2023 uninsured motorist statistics)
• Recommendation Letter Generation: Drafts E&O-safe client recommendation letters with source-cited coverage suggestions per NAIC Model #670
• Multi-Carrier Normalization: Automatically normalizes coverage terminology per ISO form taxonomy (HO-3 A–F, PAP Parts A–F, BOP sections)
• ISO Form Detection: Identifies ISO form types (HO-3, HO-5, HO-6, HO-8, PAP PP 00 01, BOP BP 00 01) from declarations pages for form-specific analysis
• Exclusion Gap Analysis: Flags high-impact exclusions (flood, earthquake, sewer backup) with specific endorsement recommendations (NFIP, HO-31, HO-46, HO-04 77)
• State-Specific Compliance: Compares auto liability limits against state minimum requirements for 10 major states per NAIC/DOI data
• E&O Liability Protection: All recommendation language follows E&O-safe patterns — never guarantees outcomes, never advises dropping coverage
• Coinsurance Penalty Detection: Warns when HO-3 dwelling coverage falls below 80% of replacement cost per ISO coinsurance clause
• Centralized Knowledge Base: Complete insurance domain reference accessible via /pc:pc-kb — ISO forms, NAIC model laws, benchmarks, and E&O guardrails

Estimated Cost per Use

Disclaimer: Token estimates are approximate and based on typical usage patterns measured from skill prompt sizes. Actual costs vary with input data size, conversation length, and complexity. Estimates use Claude Sonnet 4.6 pricing ($3/1M input, $15/1M output). Cowork and Claude Desktop subscription users (Pro/Max/Team) are not charged per-token — these estimates apply only to direct Anthropic API usage. Running stages individually in fresh sessions uses fewer input tokens than running the full pipeline sequentially, because pipeline mode accumulates conversation history across stages.

Per skill (run individually in a fresh session):

Stage	Skill Prompt	User Input	Total Input	Output	Est. Cost
pc-gap-analysis	~4.2K	~800	~7.8K	~6.0K	~$0.11
pc-kb	~3.8K	~800	~7.4K	~3.8K	~$0.08
pc-recommend	~3.8K	~800	~7.5K	~3.8K	~$0.08
pc-compare	~3.1K	~800	~6.7K	~3.1K	~$0.07
pc-ingest	~3.6K	~2.0K	~8.3K	~2.0K	~$0.05
Standalone total			~37.6K	~18.7K	~$0.39

Full pipeline (all stages in one session — context accumulates):

Stage	Base Input	+ History	Total Input	Output	Est. Cost
pc-gap-analysis	~8.0K	0	~8.0K	~6.0K	~$0.11
pc-kb	~7.6K	~6.8K	~14.4K	~3.8K	~$0.10
pc-recommend	~7.6K	~11.4K	~19.0K	~3.8K	~$0.11
pc-compare	~6.9K	~16.0K	~22.9K	~3.1K	~$0.12
pc-ingest	~8.6K	~19.9K	~28.4K	~2.0K	~$0.12
Pipeline total			~92.7K	~18.7K	~$0.56

Running the full pipeline once typically costs $0.39–$0.73 in API tokens (Claude Sonnet 4.6).

Known Limitations

1. Policy count cap: Maximum 8 policies per run. Above 8, quality degrades and hallucination risk increases. Recommended maximum is 6 per run.
2. Static benchmarks: Coverage benchmarks are sourced from III recommended levels and ISO form standards (as of v1.1). State-specific auto minimums included for 10 major states. Benchmarks do not auto-update — see /pc:pc-kb for current sources.
3. Scanned PDFs not supported: Image-only PDFs (no embedded text layer) cannot be processed. Copy-paste the text into a .txt file as a workaround.
4. Informational only: All outputs include a mandatory legal disclaimer. This tool does not constitute licensed insurance advice.
5. Single session only: Processing files live in the workspace folder on your machine. There is no cloud sync or cross-device access.

---

Context Guide

For best results, provide clear, complete policy declaration pages. The more complete the declaration page text, the more accurate the extraction. Fields not found on the declarations page are set to null (not fabricated).

If a run produces incomplete data, check processing/pc-policies.json → extraction_notes array for fields that could not be confidently extracted.

---

Legal Notice

All plugin outputs include the following mandatory disclaimer:

This letter is for informational purposes only and does not constitute professional insurance advice. Coverage terms, conditions, exclusions, and availability vary by state, carrier, and individual circumstances. This analysis is based solely on the policy declarations provided and may not reflect the complete terms of your coverage. Please consult a licensed insurance professional before making any coverage decisions.

---

Version

1.1 — Deep Domain Knowledge (ISO forms, NAIC model laws, state-specific benchmarks, E&O guardrails, centralized knowledge base)

Important Disclaimers

• AI-Generated Content: This plugin uses AI (LLM) technology which can produce inaccurate or incomplete outputs. All content should be treated as a starting point and reviewed for accuracy before use.
• Not Professional Advice: Outputs do not constitute legal, financial, tax, medical, or other professional advice. Consult qualified professionals before making decisions based on generated content.
• No Compliance Guarantee: References to industry standards, regulations, or guidelines are for informational purposes only. This plugin does not guarantee compliance with any law or regulation. Users are responsible for verifying all outputs meet their specific regulatory requirements.
• No Endorsement or Affiliation: Mention of third-party products, standards, or organizations does not imply endorsement, partnership, or certification by those entities.