abcnuts/project-handoff-ingestion

Project Handoff Ingestion

Systematic process for ingesting projects handed off from other development environments, verifying all components, and creating comprehensive deployment documentation.

When to Use This Skill

Use this skill when you receive a project handoff and need to:

• Verify access to all project components (code, databases, APIs, deployments)
• Understand the current deployment status vs. documentation claims
• Identify gaps between documented and actual system state
• Create actionable deployment guides
• Document the complete system architecture

Core Workflow

Phase 1: Initial Documentation Analysis

Read and analyze all provided handoff documentation:

• Project overview documents
• Deployment guides
• Architecture diagrams
• Environment variable lists
• API documentation

Create an initial assessment document tracking:

• Claimed system status
• Required credentials
• Deployment platforms mentioned
• Known gaps or warnings

Phase 2: Service Access Verification

Systematically verify access to each mentioned service:

Live Services (HTTP/HTTPS)

• Test health endpoints
• Verify API responses
• Check deployment status
• Record response times

Databases

• Use Supabase MCP if available
• Verify schema matches documentation
• Check for existing data
• Query table structures

Source Code Repositories

• Attempt to clone repositories
• If access denied, document the exact error
• Request collaborator access from user
• Verify repository structure matches documentation

Deployment Platforms

• Check Railway/Vercel/Netlify dashboards if accessible
• Verify environment variables
• Review deployment logs
• Check for authentication errors

Phase 3: Discrepancy Identification

Compare documentation claims against actual findings:

Create a discrepancy table: | Component | Documentation Claims | Actual Finding | Impact | |-----------|---------------------|----------------|--------| | Example | "95% complete" | "60% complete, payment system not deployed" | High |

Common discrepancies to check:

• Completion percentage vs. actual deployment
• Environment variables (claimed missing vs. actually configured)
• API endpoints (documented vs. actually deployed)
• Service integrations (documented vs. functional)

Phase 4: Code Analysis

Once repository access is obtained:

Examine Project Structure

• List all directories and key files
• Identify main application files
• Find configuration files
• Locate deployment scripts

Analyze Implementation

• Read main server/application files
• Identify all API endpoints
• Check for payment/authentication logic
• Verify notification service integrations
• Review error handling

Check Dependencies

• Read package.json/requirements.txt
• Verify all dependencies are documented
• Check for version mismatches

Phase 5: Endpoint Testing

Test all documented API endpoints:

# Health check
curl https://api-url/health

# POST endpoints
curl -X POST https://api-url/endpoint \
  -H "Content-Type: application/json" \
  -d '{"test": "data"}'

Document results:

• ✅ Working as expected
• ⚠️ Working with errors
• ❌ Not found (404)
• ❌ Server error (500)

Phase 6: Database Verification

If using Supabase MCP:

# List all tables
manus-mcp-cli tool call execute_sql --server supabase \
  --input '{"project_id":"PROJECT_ID","query":"SELECT table_name FROM information_schema.tables WHERE table_schema = '\''public'\'';"}'  

# Check table schema
manus-mcp-cli tool call execute_sql --server supabase \
  --input '{"project_id":"PROJECT_ID","query":"SELECT * FROM table_name LIMIT 5;"}'

Verify:

• Tables exist as documented
• Schema matches CREATE TABLE statements
• Sample data is present
• Data types are correct

Phase 7: Execution Log Creation

Maintain a detailed execution log throughout the process:

# Project Ingestion Log

## Phase 1: Documentation Analysis ✅
- Read handoff document
- Identified 5 services
- Found 12 environment variables

## Phase 2: Service Verification ⚠️
- API: ✅ Online (response time: 450ms)
- Database: ✅ Accessible via MCP
- Repository: ❌ Access denied
- Frontend: ✅ Deployed and responsive

## Phase 3: Discrepancies Found
1. Payment endpoints documented but not deployed
2. Stripe keys claimed missing but actually configured
...

Phase 8: Comprehensive Documentation

Create final deliverables:

System Health Report

• Executive summary
• Component status table
• Discrepancy analysis
• Actionable recommendations

Deployment Guide

• Step-by-step deployment instructions
• Corrected environment variable lists
• Testing procedures
• Troubleshooting section

GitHub Access Pattern

When repository access is denied:

1. Verify the error: Attempt clone and capture exact error message
2. Check authentication: Run gh auth status to verify GitHub CLI is authenticated
3. Request access: Ask user to add collaborator via Settings → Collaborators
4. Provide guidance: Give user step-by-step instructions with screenshots if needed
5. Alternative approach: Suggest making repository temporarily public
6. Confirm access: Once granted, clone immediately and notify user they can revert to private

Database Exploration Pattern

When exploring unknown databases:

1. List all schemas: Query information_schema for all tables
2. Check for configuration tables: Look for tables named *config*, *secret*, *setting*
3. Examine project registry: Check for project_registry or similar metadata tables
4. Query app secrets: Look for credential storage tables
5. Verify data integrity: Sample data from main tables to ensure schema matches docs

Common Pitfalls

Don't assume documentation is accurate: Always verify claims against actual system state.

Don't skip endpoint testing: Even if documentation says endpoints exist, test them.

Don't forget to log discrepancies: Track every difference between docs and reality.

Don't overlook environment variables: Check deployment platforms directly, not just documentation.

Don't proceed without repository access: Code analysis is critical for understanding implementation.

Deliverables Checklist

• [ ] Execution log with all phases documented
• [ ] System health report with component status
• [ ] Discrepancy analysis table
• [ ] Comprehensive deployment guide
• [ ] API endpoint test results
• [ ] Database schema verification
• [ ] Actionable next steps

Success Criteria

The ingestion is complete when you can answer:

• What is the actual system status (not claimed status)?
• What works and what doesn't?
• What are the exact steps to deploy missing components?
• What credentials are needed and where to get them?
• What are the known issues and how to fix them?