Skip to content

Troubleshooting Guide

Solutions to common AEL issues.

Installation Issues

Command Not Found

zsh: command not found: ael

Solutions: 1. Ensure AEL is installed: pip install ael 2. Check PATH includes pip bin directory 3. Try: python -m ael --version

Import Errors

ModuleNotFoundError: No module named 'ael'

Solutions: 1. Verify installation: pip show ael 2. Check Python version (requires 3.11+) 3. Reinstall: pip install --force-reinstall ael

Configuration Issues

Config File Not Found

Error: Configuration file not found

Solutions: 1. Create config: ael config show > ael-config.yaml 2. Specify path: ael --config /path/to/config.yaml 3. Check current directory for ael-config.yaml

Invalid Configuration

Error: CONFIG_INVALID - Invalid configuration

Solutions: 1. Validate YAML syntax 2. Check required fields 3. View current config: ael config show 4. See Configuration Reference

Workflow Issues

Workflow Not Found

Error: WORKFLOW_NOT_FOUND - Workflow 'my-workflow' not found

Solutions: 1. List workflows: ael workflows list 2. Check workflows directory in config 3. Verify file exists and has .yaml extension 4. Check workflow name field matches

Validation Errors

Error: Workflow validation failed

Solutions: 1. Validate workflow: ael validate workflow.yaml 2. Check required fields: name, version, steps 3. Verify step IDs are unique 4. Check depends_on references valid steps

Circular Dependency

Error: CIRCULAR_DEPENDENCY - Circular dependency detected

Solutions: 1. Review depends_on relationships 2. Draw dependency graph 3. Remove circular references 4. Use sequential steps without depends_on

Tool Issues

Tool Unavailable

Error: TOOL_UNAVAILABLE - Tool 'tool_name' is unavailable

Solutions: 1. List tools: ael tools list 2. Check MCP server is configured 3. Refresh tools: ael tools refresh 4. Verify MCP server is running

MCP Connection Failed

Error: MCP_CONNECTION_FAILED - Failed to connect to MCP server

Solutions: 1. Check server command: npx -y @package/server 2. Verify package is installed 3. Check environment variables 4. Review server logs

Tool Timeout

Error: TOOL_TIMEOUT - Tool timed out after 30s

Solutions: 1. Increase step timeout:

steps:
  - id: slow_step
    tool: slow_tool
    timeout: 120
2. Check tool is not stuck 3. Verify network connectivity

Code Step Issues

Syntax Error

Error: CODE_SYNTAX - Syntax error in code block

Solutions: 1. Check Python syntax 2. Verify indentation (use spaces, not tabs) 3. Test code locally first 4. Check for unclosed brackets/quotes

Security Violation

Error: CODE_SECURITY - Security violation in code block

Solutions: 1. Remove forbidden imports (os, sys, subprocess) 2. Don't use eval(), exec(), open() 3. See Code Steps Guide for allowed imports

Import Not Allowed

Error: Import 'requests' not allowed

Solutions: 1. Use allowed imports only (json, re, datetime, etc.) 2. Use tool steps for HTTP requests 3. See Code Steps Guide

Result Not Set

Error: Code step did not set 'result' variable

Solutions: 1. Ensure code sets result:

code: |
  data = process()
  result = data  # Required!

Execution Issues

Workflow Timeout

Error: WORKFLOW_TIMEOUT - Workflow timed out after 300s

Solutions: 1. Increase workflow timeout in config 2. Optimize slow steps 3. Check for stuck tools 4. Break into smaller workflows

Template Error

Error: TEMPLATE_ERROR - Template rendering failed

Solutions: 1. Check Jinja2 syntax: {{ variable }} 2. Verify variable names exist 3. Check for typos in step/input names 4. Use | default('value') for optional values

Server Issues

Port Already in Use

Error: Address already in use (port 8080)

Solutions: 1. Use different port: ael serve --port 8081 2. Find process: lsof -i :8080 3. Kill process: kill -9 <PID>

API Authentication Failed

Error: 401 Unauthorized

Solutions: 1. Check API key is configured 2. Verify key in request header 3. Check security config in ael-config.yaml

Getting Help

Debug Logging

Enable debug logs:

ael --log-level DEBUG run workflow.yaml

Or in config:

logging:
  level: DEBUG

Check Version

ael version

Report Issues

  1. Gather logs with DEBUG level
  2. Include workflow YAML (redact secrets)
  3. Include error message
  4. Open issue on GitHub