worldarchitect-local-debugging

name	worldarchitect-local-debugging
description	Guide and workflow for WorldArchitect.AI Local Debugging Reference. Use when you need WorldArchitect.AI Local Debugging Reference.

WorldArchitect.AI Local Debugging Reference

Log File Locations

Flask server logs:

/tmp/worldarchitect.ai/<branch-name>/flask-server.log

MCP server logs:

/tmp/worldarchitect.ai/<branch-name>/mcp-server.log

Quick Log Commands

# Tail recent logs for current branch
tail -100 /tmp/worldarchitect.ai/codex_integrate-gemini-and-openrouter-apis-czdlzo/flask-server.log

# Search for errors
grep -i "error\|cerebras\|400\|500" /tmp/worldarchitect.ai/*/flask-server.log | tail -50

# Watch logs in real-time
tail -f /tmp/worldarchitect.ai/*/flask-server.log

When Server Restart is Required

Python caches modules, so restart needed after changing:

• constants.py - any constant values
• Any __init__.py files
• Import structure changes
• New dependencies

No restart needed for:

• Template changes (Jinja reloads)
• Static files (CSS/JS)
• Most Flask route logic (with debug mode)

Running Server

# From project root
./run_local_server.sh

# Server runs on port 8005-8007 (check output)

Common Debug Patterns

# Find provider/model being used
grep "MODEL_INFO\|Using provider" /tmp/worldarchitect.ai/*/flask-server.log | tail -10

# Find token/context errors
grep "context_length\|token" /tmp/worldarchitect.ai/*/flask-server.log | tail -10

# Find API errors with response body
grep -A2 "API error" /tmp/worldarchitect.ai/*/flask-server.log | tail -20