/**
* Document Compliance Processor
*
* This script provides functionality for autonomous regulatory monitoring of documents
* using a ReAct agent and Document Engine MCP Server. It performs the following operations:
* - Scans documents for regulatory adherence
* - Automatically applies redactions for sensitive data
* - Creates audit trails
* - Generates compliance reports
* - Monitors document repositories for policy violations
* - Applies protective measures
* - Maintains comprehensive compliance documentation
*
* Usage: node document-processor.js <path/to/file>
*/
import { createReactAgent } from '@langchain/langgraph/prebuilt';
import { config } from 'dotenv';
import {
createMCPClient,
createLLM,
parseDocumentIds,
handleAgentStream,
processCommandLine,
uploadDocumentWorkflow,
} from './shared/common.js';
import { uploadDocuments, downloadFiles } from './shared/fileHandlers.js';
config();
async function processCompliance(filePaths) {
console.log(`🛡️ Starting compliance processing for ${filePaths.length} documents...`);
const documentIds = await uploadDocumentWorkflow(
filePaths,
uploadDocuments,
'Compliance processing'
);
// Connect to Document Engine MCP Server
const { tools } = await createMCPClient();
// Create LLM model
const llm = createLLM({ model: 'gpt-4.1-mini', temperature: 0 });
// Create ReAct agent with Document Engine tools
const agent = createReactAgent({
llm: llm,
tools: tools,
}).withConfig({
recursionLimit: 100,
});
// Give agent the compliance processing task with streaming
const systemMessage = `You are a compliance monitoring agent executing in a tool-augmented environment.
Your task: Immediately begin scanning document(s): ${documentIds.join(', ')} for regulatory compliance issues using the provided Document Engine tools.
⚠️ CRITICAL INSTRUCTION: DO NOT NARRATE your plan or write summaries. Begin directly with the first tool call. No explanation. No reasoning first. Just take action using tools.
If you output anything other than a tool call, the system will assume your job is finished and terminate.
---
Your workflow:
1. **Comprehensive PII Protection**
- Identify and redact ALL types of personally identifiable information including:
* First, Middle and last names
* Identity numbers (SSN, driver's license, passport, etc.)
* Financial information (credit cards, bank accounts, etc.)
* Contact details (names, addresses, phone numbers, emails)
* Location data (addresses, ZIP codes, coordinates)
* Tax related numbers
* Dates and times that could identify individuals
* Any other sensitive personal data
2. **Document Security Enhancement**
- Apply appropriate confidentiality classifications and watermarks
- Add compliance annotations documenting protection measures taken
- Create comprehensive audit trail of all security actions
3. **Multi-Stakeholder Versions**
- Generate appropriately redacted versions for different access levels
- Ensure sensitive data is properly protected while maintaining document utility
- Create versions suitable for public disclosure, internal review, and audit purposes
4. **Compliance Verification**
- Ensure all regulatory requirements are met (GDPR, CCPA, HIPAA, etc.)
- Document any areas requiring manual legal review
- Verify completeness of data protection measures
---
🚨 Repeat: NO explanations. Start with the first tool call.
✅ When fully complete, output the final result in this exact format:
**FINAL_DOCUMENT_IDS: [document_id_1, document_id_2, ...]**
Replace with actual Document Engine IDs.
`;
console.log('🤖 Starting document compliance processing with ReAct agent...');
// Stream the agent's execution
const stream = await agent.stream(
{
messages: [
{ role: 'system', content: systemMessage },
{
role: 'user',
content:
'Please scan these documents for compliance issues and apply necessary protections.',
},
],
},
{
streamMode: 'values',
}
);
const finalResponse = await handleAgentStream(stream);
console.log('✅ Compliance processing complete!');
const finalDocumentIds = parseDocumentIds(finalResponse);
console.log('📄 Final Document IDs:', finalDocumentIds);
return finalDocumentIds;
}
await processCommandLine(async filePaths => {
const documentIds = await processCompliance(filePaths);
await downloadFiles(documentIds);
}, 'Usage: node document-processor.js <path/to/file>');
