Debugging Tasks

If a task fails or behaves unexpectedly, you can debug it directly from the Anchor Automate dashboard.

---

Step-by-Step Debugging

1. Go to Tasks

   • Navigate to the "Tasks" tab from the left sidebar.

2. Select the Failed Task

   • Filter by failed status to find problematic tasks quickly.

3. Inspect Step Logs

   • Each step has its own logs with detailed output.
   • Look for error messages, stack traces, or invalid inputs.

4. Check Input & Output

   • Review input parameters passed to each step.
   • Compare expected output with actual result.
   • Fixes can often be made by correcting the input parameters in the Task Definition.

5. Enable Retry or Edit

   • You can retry the task or go back and edit your Task Definition to fix the logic.

---

Debugging with Agents

tip

You don’t have to do it all manually, Agents can help!

• Ask the Agent

  • If a task fails, simply ask the Agent what went wrong.
  • Agents can analyze logs, context, and errors to provide insights.

• Guide the Agent

  • After the Agent points out the issue, you can offer the solution directly.
  • Solutions may include updating input parameters, fixing logic in steps, or correcting configuration settings.
  • This keeps the debugging process collaborative

---

Common Debugging Tips

• Use console logs inside custom code blocks.
• Verify third-party app credentials (e.g., API keys).
• Test each step in isolation before chaining.
• Check input parameters in the Task Definition.

---

Debug Mode (Coming Soon)

A sandboxed debug mode for dry-running task definitions is in the works! 🛠️