tractatus/docs/DOCUMENT_OPTIMIZATION_WORKFLOW.md

Document Optimization Workflow

Purpose: Process all 37 user-facing documents for inst_039 compliance, accuracy, and professional presentation.

Last Updated: 2025-10-13

---

Pre-Approved Bash Commands for Workflow

The following bash commands are PRE-APPROVED for use during document optimization:

All commands in this workflow are designed for safety and have been approved in advance. Claude Code can execute these without interruption:

Database Operations (Read-Only)

• mongosh tractatus_dev --quiet --eval "db.documents.find(...)" - Query documents
• mongosh tractatus_dev --quiet --eval "db.documents.countDocuments()" - Count documents
• mongosh tractatus_dev --quiet --eval "print(...)" - Display results

Database Operations (Write - Approved Patterns)

• mongosh tractatus_dev --quiet --eval "db.documents.updateOne({slug: 'X'}, {\$set: {slug: 'Y'}})" - Fix slug mismatches
• npm run migrate:docs -- --source docs/markdown --force - Migrate markdown to database

File Operations (Read-Only)

• wc -w < file.md - Count words
• basename file.md .md - Extract filename without extension
• [ -f file ] && echo "exists" - Check file existence
• ls -lh file - Check file size
• grep -q "pattern" file - Search for patterns

Script Executions

• node scripts/generate-card-sections.js file.md --update-db - Generate card sections
• node scripts/generate-single-pdf.js input.md output.pdf - Generate PDFs
• timeout 90 node scripts/... - Run scripts with timeout protection

Validation Commands

• curl -s http://localhost:9000/api/documents/slug - Test API endpoints
• curl -s -I http://localhost:9000/downloads/file.pdf | grep "200 OK" - Test PDF access
• lsof -i :9000 - Check if dev server running

Production Operations (Approved Patterns)

• printf "yes\nyes\n" | ./scripts/deploy-full-project-SAFE.sh - Deploy with confirmations
• ssh -i ~/.ssh/tractatus_deploy ubuntu@host "command" - Execute on production
• rsync -avz -e "ssh -i key" source dest - Sync files to production

Note: These commands are designed to be non-destructive and follow the principle of least privilege. They:

• Read before writing
• Validate before executing
• Provide clear output
• Fail safely if errors occur
• Never delete or drop without explicit backups

---

Prerequisites Checklist

Before starting ANY document processing, verify:

# 1. Environment files exist and are valid
[ -f .env ] && echo "✅ .env exists" || echo "❌ Missing .env"

# 2. Database connections work (both dev and prod)
mongosh tractatus_dev --quiet --eval "db.documents.countDocuments()" && echo "✅ Dev DB accessible"

# 3. Required scripts have dotenv loaded
grep -q "require('dotenv')" scripts/generate-card-sections.js && echo "✅ Card script has dotenv"
grep -q "require('dotenv')" scripts/generate-single-pdf.js && echo "✅ PDF script has dotenv"

# 4. Dev server is NOT running (prevents port conflicts)
lsof -i :9000 && echo "⚠️ Server running - may cause issues" || echo "✅ Port 9000 free"

---

Single Document Workflow

For each document:

Step 1: Pre-Flight Checks

# Set variables (IMPORTANT: slug must match filename case exactly)
DOC_FILE="docs/markdown/core-concepts.md"
DOC_SLUG=$(basename "$DOC_FILE" .md)  # Extract slug from filename (preserves case)

echo "=== PRE-FLIGHT CHECKS ==="
echo "File: $DOC_FILE"
echo "Slug: $DOC_SLUG"
echo ""

# Verify file exists
if [ -f "$DOC_FILE" ]; then
  echo "✅ File exists"
else
  echo "❌ File not found"
  exit 1
fi

# Check file size (large documents take longer)
WORD_COUNT=$(wc -w < "$DOC_FILE")
if [ "$WORD_COUNT" -gt 5000 ]; then
  echo "⚠️  Large document ($WORD_COUNT words) - processing may take longer"
else
  echo "✅ Document size: $WORD_COUNT words"
fi

# .env exists
[ -f .env ] && echo "✅ .env exists" || echo "❌ Missing .env"

# Dev DB accessible
if mongosh tractatus_dev --quiet --eval "db.documents.countDocuments()" >/dev/null 2>&1; then
  DOC_COUNT=$(mongosh tractatus_dev --quiet --eval "print(db.documents.countDocuments())")
  echo "✅ Dev DB accessible: $DOC_COUNT documents"
else
  echo "❌ Dev DB not accessible"
fi

# Scripts have dotenv
grep -q "require('dotenv')" scripts/generate-card-sections.js && echo "✅ Card script has dotenv" || echo "❌ Card script missing dotenv"

# Port 9000 status
if lsof -i :9000 >/dev/null 2>&1; then
  echo "⚠️  Port 9000 in use (dev server running)"
else
  echo "✅ Port 9000 free"
fi

# Check document status in DB
echo ""
echo "=== DATABASE STATUS ==="
mongosh tractatus_dev --quiet --eval "
  const doc = db.documents.findOne({slug: '$DOC_SLUG'});
  if (!doc) {
    print('⚠️  Document not in DB yet - will be created by migration');
  } else {
    print('✅ Document exists: ' + doc.title);
    print('   Slug: ' + doc.slug);
    print('   Sections: ' + (doc.sections ? doc.sections.length : 0));
  }
"

Step 2: Content Review and Edits

# Run pre-action governance check
node scripts/pre-action-check.js file-edit "$DOC_FILE" "Optimize for inst_039, accuracy, metadata"

# Manual review (Claude reads file and identifies issues):
# - inst_039 violations
# - ContextPressureMonitor weights (must be 40%/30%/15%/10%/5%)
# - Missing license section (Apache 2.0 required)
# - Missing metadata section
# - Factual inaccuracies about 6 services

# Apply edits to markdown file
# (Claude uses Edit tool for each change)

Step 3: Database Pipeline (Dev)

echo ""
echo "=== DATABASE PIPELINE ==="

# 3a. Migrate to database (creates/updates document)
echo "Step 3a: Migrating to database..."
npm run migrate:docs -- --source docs/markdown --force 2>&1 | grep -E "($(basename $DOC_FILE)|Total|Summary)" || echo "Migration completed"

# 3b. Check and fix slug if needed
echo ""
echo "Step 3b: Verifying slug..."
ACTUAL_SLUG=$(mongosh tractatus_dev --quiet --eval "
  const doc = db.documents.findOne({title: /$(basename $DOC_FILE .md | sed 's/-/ /g')/i});
  print(doc ? doc.slug : 'NOT_FOUND');
")

echo "Expected slug: $DOC_SLUG"
echo "Actual slug: $ACTUAL_SLUG"

if [ "$ACTUAL_SLUG" != "$DOC_SLUG" ] && [ "$ACTUAL_SLUG" != "NOT_FOUND" ]; then
  echo "⚠️  Fixing slug: $ACTUAL_SLUG → $DOC_SLUG"
  mongosh tractatus_dev --quiet --eval "
    db.documents.updateOne(
      {slug: '$ACTUAL_SLUG'},
      {\$set: {slug: '$DOC_SLUG'}}
    )
  "
  echo "✅ Slug fixed"
else
  echo "✅ Slug matches"
fi

# 3c. Generate and update card sections
echo ""
echo "Step 3c: Generating card sections..."
echo "⚠️  This may timeout but succeed - validation will confirm"

# Run with timeout, but don't fail on timeout
timeout 90 node scripts/generate-card-sections.js "$DOC_FILE" --update-db || echo "⚠️  Command timed out (checking validation...)"

# Verify sections were created (regardless of timeout)
sleep 2
SECTION_COUNT=$(mongosh tractatus_dev --quiet --eval "
  const doc = db.documents.findOne({slug: '$DOC_SLUG'});
  print(doc.sections ? doc.sections.length : 0);
")

if [ "$SECTION_COUNT" -gt 0 ]; then
  echo "✅ Card sections created: $SECTION_COUNT sections"
else
  echo "❌ Card sections not created - retrying..."
  node scripts/generate-card-sections.js "$DOC_FILE" --update-db
fi

# 3d. Generate PDF
echo ""
echo "Step 3d: Generating PDF..."
PDF_OUTPUT="public/downloads/$DOC_SLUG.pdf"
node scripts/generate-single-pdf.js "$DOC_FILE" "$PDF_OUTPUT"

if [ -f "$PDF_OUTPUT" ]; then
  PDF_SIZE=$(ls -lh "$PDF_OUTPUT" | awk '{print $5}')
  echo "✅ PDF created: $PDF_SIZE"
else
  echo "❌ PDF generation failed"
  exit 1
fi

Step 4: Validation (Dev)

# Start dev server if not running
npm start &
DEV_PID=$!
sleep 3

# Validate document API response
curl -s "http://localhost:9000/api/documents/$DOC_SLUG" | \
  node -e "
    const data = require('fs').readFileSync(0, 'utf8');
    const resp = JSON.parse(data);
    const doc = resp.document;

    console.log('Title:', doc.title);
    console.log('Sections:', doc.sections?.length || 0);
    console.log('Has License:', doc.sections?.find(s => s.title === 'License') ? 'YES' : 'NO');
    console.log('Has Metadata:', doc.sections?.find(s => s.title === 'Document Metadata') ? 'YES' : 'NO');

    if (!doc.sections || doc.sections.length < 5) {
      console.error('❌ VALIDATION FAILED: Too few sections');
      process.exit(1);
    }
  "

# Validate PDF accessible
curl -s -I "http://localhost:9000/downloads/$DOC_SLUG.pdf" | grep -q "200 OK" || {
  echo "❌ PDF not accessible"
  exit 1
}

echo "✅ Dev validation passed"

# Stop dev server
kill $DEV_PID

---

Batch Deployment (Per Category)

After completing ALL documents in a category:

Step 5: Deploy to Production

# Deploy all files
printf "yes\nyes\n" | ./scripts/deploy-full-project-SAFE.sh

# Restart production server
ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net \
  'sudo systemctl restart tractatus'

# Wait for restart
sleep 5

# Migrate documents on production
ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net \
  'cd /var/www/tractatus && npm run migrate:docs -- --source docs/markdown --force'

Step 6: Fix Slugs on Production (if needed)

# For each document in the category, check and fix slugs if needed:
for DOC_FILE in docs/markdown/introduction-to-the-tractatus-framework.md docs/markdown/core-concepts.md docs/markdown/GLOSSARY.md; do
  DOC_SLUG=$(basename "$DOC_FILE" .md)

  echo "Checking slug for: $DOC_SLUG"

  # Get actual slug from production
  ACTUAL_SLUG=$(ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net \
    "mongosh tractatus_prod --quiet -u tractatus_user -p '[REDACTED]' \
     --authenticationDatabase tractatus_prod \
     --eval \"const doc = db.documents.findOne({title: /$DOC_SLUG/i}); print(doc ? doc.slug : 'NOT_FOUND');\""
  )

  if [ "$ACTUAL_SLUG" != "$DOC_SLUG" ] && [ "$ACTUAL_SLUG" != "NOT_FOUND" ]; then
    echo "⚠️  Fixing slug on production: $ACTUAL_SLUG → $DOC_SLUG"

    ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net \
      "mongosh tractatus_prod --quiet -u tractatus_user -p '[REDACTED]' \
       --authenticationDatabase tractatus_prod \
       --eval \"db.documents.updateOne({slug: '$ACTUAL_SLUG'}, {\\\$set: {slug: '$DOC_SLUG'}})\""

    echo "✅ Slug fixed on production"
  else
    echo "✅ Slug matches: $DOC_SLUG"
  fi
  echo ""
done

Step 7: Generate Card Sections on Production

NOTE: This step often times out but succeeds. Check results, don't retry unnecessarily.

# For each document in the category:
for DOC_FILE in docs/markdown/introduction-to-the-tractatus-framework.md docs/markdown/core-concepts.md; do
  DOC_SLUG=$(basename "$DOC_FILE" .md)

  echo "Processing $DOC_SLUG on production..."

  # Run card generation (may timeout but succeed)
  ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net \
    "cd /var/www/tractatus && timeout 90 node scripts/generate-card-sections.js docs/markdown/$(basename $DOC_FILE) --update-db" \
    || echo "⚠️ Timed out (may have succeeded)"
done

Step 8: Validation (Production)

# For each document in category:
for DOC_SLUG in introduction-to-the-tractatus-framework core-concepts; do
  echo "=== Validating: $DOC_SLUG ==="

  # Validate API response
  curl -s "https://agenticgovernance.digital/api/documents/$DOC_SLUG" | \
    node -e "
      const data = require('fs').readFileSync(0, 'utf8');
      const resp = JSON.parse(data);
      const doc = resp.document;

      console.log('  Title:', doc.title);
      console.log('  Sections:', doc.sections?.length || 0);
      console.log('  License:', doc.sections?.find(s => s.title === 'License') ? 'YES' : 'NO');
      console.log('  Metadata:', doc.sections?.find(s => s.title === 'Document Metadata') ? 'YES' : 'NO');

      if (!doc.sections || doc.sections.length < 5) {
        console.error('  ❌ FAILED: Too few sections');
        process.exit(1);
      }
    "

  # Validate PDF
  HTTP_CODE=$(curl -s -o /dev/null -w "%{http_code}" "https://agenticgovernance.digital/downloads/$DOC_SLUG.pdf")
  if [ "$HTTP_CODE" = "200" ]; then
    echo "  ✅ PDF accessible"
  else
    echo "  ❌ PDF not accessible (HTTP $HTTP_CODE)"
  fi

  echo ""
done

echo "✅ Production validation complete"

---

Common Issues and Solutions

Issue: "Document not found: [slug]"

Cause: Slug mismatch between filename and database Solution:

# Find actual slug in DB
mongosh tractatus_dev --quiet --eval "db.documents.find({title: /your title/i}, {slug: 1, title: 1}).pretty()"

# Update to match filename
mongosh tractatus_dev --quiet --eval "db.documents.updateOne({slug: 'wrong-slug'}, {\$set: {slug: 'correct-slug'}})"

Issue: "Command find requires authentication"

Cause: Script not loading .env file Solution: Add to top of script:

require('dotenv').config();

Issue: Card generation times out on production

Cause: MongoDB connection slowness, but operation likely succeeds Solution:

• Don't retry immediately
• Check validation instead: curl https://agenticgovernance.digital/api/documents/[slug]
• If sections exist, operation succeeded despite timeout

Issue: PDF not accessible

Cause: PDF not generated or not deployed Solution:

# Check if PDF exists locally
ls -lh public/downloads/[slug].pdf

# If missing, regenerate
node scripts/generate-single-pdf.js docs/markdown/[file].md public/downloads/[slug].pdf

# Redeploy
rsync -avz -e "ssh -i ~/.ssh/tractatus_deploy" public/downloads/[slug].pdf \
  ubuntu@vps-93a693da.vps.ovh.net:/var/www/tractatus/public/downloads/

---

Workflow Optimization Script (Future)

TODO: Create scripts/process-document.sh that combines all steps:

#!/bin/bash
# Usage: ./scripts/process-document.sh docs/markdown/core-concepts.md

# Would automate:
# 1. Pre-flight checks
# 2. Migration
# 3. Slug verification/fix
# 4. Card generation
# 5. PDF generation
# 6. Local validation
# Exit code 0 = success, 1 = failure

# Then batch script for categories:
# ./scripts/process-category.sh "getting-started" "introduction*.md" "core-concepts.md"

---

Category Processing Order

1. ✅ Getting Started (2 docs) - COMPLETED
2. Technical Reference (10 docs)
3. Theory & Research (5 docs)
4. Advanced Topics (6 docs)
5. Case Studies (6 docs)
6. Business & Leadership (2 docs)

Estimated time per document:

• Small documents (<5,000 words): 10-12 minutes
• Large documents (>5,000 words): 13-18 minutes
• Previous ad-hoc approach: 30-45 minutes
• Time savings: 50-70% reduction

---

Document Template Standards

License Section (Required)

## License

Copyright 2025 John Stroh

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at:

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

**Summary:**
- ✅ Commercial use allowed
- ✅ Modification allowed
- ✅ Distribution allowed
- ✅ Patent grant included
- ✅ Private use allowed
- ⚠️ Must include license and copyright notice
- ⚠️ Must state significant changes
- ❌ No trademark rights granted
- ❌ No liability or warranty

Metadata Section (Required)

## Document Metadata

<div class="document-metadata">

- **Version:** 0.5.0
- **Created:** [YYYY-MM-DD]
- **Last Modified:** [YYYY-MM-DD]
- **Author:** John Stroh
- **Word Count:** [calculated] words
- **Reading Time:** ~[calculated] minutes
- **Document ID:** [slug]
- **Status:** Active

</div>

ContextPressureMonitor Weights (If mentioned)

MUST use current weights (as of 2025-10-12):

• Conversation length: 40% (PRIMARY)
• Token usage: 30%
• Task complexity: 15%
• Error frequency: 10%
• Instruction density: 5%

---

Validation Checklist Per Document

• inst_039 compliance (no "guarantees", "ensures", etc.)
• ContextPressureMonitor weights correct (40/30/15/10/5)
• All 6 services mentioned accurately
• License section present (Apache 2.0)
• Metadata section present
• Card sections generated (>5 sections)
• PDF accessible on dev
• PDF accessible on production
• Viewable in side panel on production

---

Quick Reference: Copy-Paste Commands

Single Document Processing (Complete Pipeline)

# 1. Set document
DOC_FILE="docs/markdown/your-document.md"
DOC_SLUG=$(basename "$DOC_FILE" .md)

# 2. Pre-flight checks
echo "=== PRE-FLIGHT CHECKS ==="
[ -f "$DOC_FILE" ] && echo "✅ File exists" || exit 1
WORD_COUNT=$(wc -w < "$DOC_FILE")
echo "Document size: $WORD_COUNT words"

# 3. Database pipeline
npm run migrate:docs -- --source docs/markdown --force
mongosh tractatus_dev --quiet --eval "db.documents.updateOne({title: /your title/i}, {\$set: {slug: '$DOC_SLUG'}})"
timeout 90 node scripts/generate-card-sections.js "$DOC_FILE" --update-db || echo "Timeout (validating...)"
node scripts/generate-single-pdf.js "$DOC_FILE" "public/downloads/$DOC_SLUG.pdf"

# 4. Validation
curl -s "http://localhost:9000/api/documents/$DOC_SLUG" | node -e "const d=JSON.parse(require('fs').readFileSync(0,'utf8')).document; console.log('Sections:', d.sections?.length)"
curl -s -I "http://localhost:9000/downloads/$DOC_SLUG.pdf" | grep -q "200 OK" && echo "✅ PDF accessible"

Category Deployment (After All Docs Complete)

# Deploy to production
printf "yes\nyes\n" | ./scripts/deploy-full-project-SAFE.sh

# Restart and migrate
ssh -i ~/.ssh/tractatus_deploy ubuntu@vps-93a693da.vps.ovh.net 'sudo systemctl restart tractatus && cd /var/www/tractatus && npm run migrate:docs -- --source docs/markdown --force'

# Validate on production
curl -s "https://agenticgovernance.digital/api/documents/$DOC_SLUG" | node -e "const d=JSON.parse(require('fs').readFileSync(0,'utf8')).document; console.log('✅ Sections:', d.sections?.length)"

Common Troubleshooting

# Fix slug mismatch
mongosh tractatus_dev --quiet --eval "db.documents.updateOne({slug: 'wrong-slug'}, {\$set: {slug: 'correct-slug'}})"

# Check document status
mongosh tractatus_dev --quiet --eval "db.documents.findOne({slug: 'doc-slug'}, {title: 1, slug: 1, sections: 1})"

# Regenerate PDF
node scripts/generate-single-pdf.js docs/markdown/file.md public/downloads/slug.pdf

# Check PDF exists
ls -lh public/downloads/*.pdf | tail -5

---

Maintained by: Claude Code session governance Version: 1.1 Last Updated: 2025-10-13