{"skill":{"slug":"document-congruence-checker","displayName":"Document Congruence Checker","summary":"Compare multiple documents for congruency and consistency. Use when: User needs to verify that multiple documents (contracts, reports, forms, specs) contain...","description":"---\nname: document-congruence-checker\ndescription: |\n  Compare multiple documents for congruency and consistency.\n  Use when: User needs to verify that multiple documents (contracts, reports, forms, specs) \n  contain consistent information, identify conflicting data points, or ensure alignment \n  across document sets.\n---\n\n# Document Congruence Checker\n\n## Purpose\n\nCompare 2+ documents to identify:\n- Conflicting information (dates, names, amounts, terms)\n- Missing fields in some documents\n- Inconsistent terminology or values\n- Version mismatches\n\n## Input\n\nUser provides:\n- 2 or more documents (PDF, DOCX, TXT, MD, JSON, XML)\n- Optional: specific fields to focus on\n- Optional: tolerance rules (e.g., date format differences OK)\n\n## Process\n\n### 1. Document Parsing\n- Extract text and structured data from each document\n- Identify document type and key fields\n- Normalize data formats (dates, currencies, names)\n\n### 2. Field Mapping\n- Create unified field schema across all documents\n- Map equivalent fields (e.g., \"Client Name\" = \"Customer Name\" = \"Party A\")\n- Note fields present in some docs but not others\n\n### 3. Congruence Analysis\nFor each mapped field:\n- **Congruent**: Values match exactly (or within tolerance)\n- **Incongruent**: Values conflict\n- **Missing**: Field absent in one or more documents\n- **Ambiguous**: Values unclear or unparseable\n\n### 4. Report Generation\nOutput a discrepancy table with:\n| Field | Doc A | Doc B | Doc C | Status | Notes |\n|-------|-------|-------|-------|--------|-------|\n\n## Output Format\n\n```markdown\n## 📊 Document Congruence Report\n\n**Documents Compared:** {list}\n**Analysis Date:** {date}\n**Overall Congruence:** {High/Medium/Low} - {X}% fields aligned\n\n### ✅ Congruent Fields ({count})\n| Field | Value | Present In |\n|-------|-------|------------|\n| {field} | {value} | All docs |\n\n### ⚠️ Incongruent Fields ({count})\n| Field | Doc A | Doc B | Doc C | Discrepancy Type |\n|-------|-------|-------|-------|------------------|\n| Date | 2026-04-01 | 2026-04-10 | 2026-04-01 | Value mismatch |\n| Amount | $10,000 | $10,000 | - | Missing in Doc C |\n\n### 📋 Missing Fields ({count})\n| Field | Missing In | Impact |\n|-------|------------|--------|\n| Signature | Doc B | High - legal validity |\n\n### 🔍 Recommendations\n1. {action item}\n2. {action item}\n```\n\n## Congruence Levels\n\n| Level | Criteria |\n|-------|----------|\n| **High** | 90%+ fields congruent, no critical conflicts |\n| **Medium** | 70-89% congruent, minor conflicts only |\n| **Low** | <70% congruent, or critical conflicts present |\n\n## Critical vs Non-Critical Fields\n\n**Critical** (incongruence = high risk):\n- Dates (effective, expiry, delivery)\n- Amounts (prices, fees, penalties)\n- Names (parties, signatories)\n- Legal terms (liability, termination)\n- Identifiers (contract #, account #)\n\n**Non-Critical** (incongruence = low risk):\n- Formatting differences\n- Typos in non-legal text\n- Optional fields\n- Descriptive language variations\n\n## Usage Examples\n\n### Example 1: Contract Comparison\n```\nUser: Compare these 3 contract versions for inconsistencies\nAgent: [parses docs, outputs congruence table]\n```\n\n### Example 2: Financial Report Alignment\n```\nUser: Check if Q1 report matches Q2 report for opening balances\nAgent: [compares specific fields, flags mismatches]\n```\n\n### Example 3: Form Data Validation\n```\nUser: Verify all application forms have consistent applicant info\nAgent: [cross-checks names, IDs, dates across forms]\n```\n\n## Tools Used\n\n- `read` - Parse document contents\n- `exec` - Run document parsing tools (pdftotext, docx2txt, etc.)\n- `write` - Generate congruence report\n\n## Limitations\n\n- Scanned PDFs require OCR (may have accuracy issues)\n- Handwritten content not reliably parseable\n- Semantic equivalence (same meaning, different words) requires human judgment\n- Images/charts not analyzed for data congruence\n\n## Security Notes\n\n- Never send sensitive documents to external APIs\n- Process locally when possible\n- Redact confidential data from reports if sharing\n- Delete temp files after analysis\n\n---\n\n*Skill ready for document congruence analysis*\n","topics":["Document"],"tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":355,"installsAllTime":13,"installsCurrent":1,"stars":0,"versions":1},"createdAt":1778001266451,"updatedAt":1778492850251},"latestVersion":{"version":"1.0.0","createdAt":1778001266451,"changelog":"Version bump","license":"MIT-0"},"metadata":null,"owner":{"handle":"dorjenorbulim","userId":"s174h6jf4cmxs108b7dcwrwjbd847qva","displayName":"dorjenorbulim","image":"https://avatars.githubusercontent.com/u/269761677?v=4"},"moderation":null}