Auto-Debug Command Execution

This skill automates the process of executing a shell command, monitoring its output for errors, attempting to debug and fix identified issues, and re-running the command until it succeeds.

Workflow

When this skill is activated with a shell command, follow these steps:

1. Execute Command (Detached Mode):

   • Run the provided command using run_shell_command in detached mode (-d) if applicable (e.g., docker-compose up --build -d). This prevents the command from blocking the agent's execution.
   • If the command is not a long-running process and doesn't support detached mode, execute it directly and capture its output.

2. Monitor and Collect Logs:

   • Immediately after executing the command, use appropriate tools to collect logs. For Docker Compose, this means docker-compose logs --no-log-prefix. For other commands, analyze the direct output or relevant log files.

3. Analyze Logs for Errors:

   • Scan the collected logs for keywords indicating errors (e.g., ERROR, Failed, Exception, cannot, not found).
   • Identify the most recent or critical error message and its context (e.g., file path, line number, class name).

4. Diagnose and Propose Fix (Internal Reasoning):

   • Based on the identified error, diagnose the root cause.
   • Formulate a concrete plan to fix the error. This may involve:
     • Modifying configuration files (write_file or replace).
     • Renaming files (run_shell_command mv).
     • Adjusting environment variables.
     • Updating Dockerfile contents.
     • Consulting internal knowledge (e.g., OpenIG configuration patterns).
   • Crucially: If the fix involves code/config modification, ensure it adheres to project conventions and existing patterns.

5. Apply Fix:

   • Execute the necessary tool calls (e.g., write_file, replace, run_shell_command) to apply the proposed fix.

6. Clean Up (if necessary):

   • If the command involves Docker containers, always bring them down (docker-compose down) before attempting to re-run docker-compose up --build -d with a new configuration. This ensures a clean state.

7. Loop or Conclude:

   • Go back to Step 1 (Execute Command) and repeat the cycle until the logs indicate successful completion of the initial command without critical errors.
   • If a series of attempts (e.g., 3-5 iterations) fails to resolve the issue, or if the error seems unresolvable given current tools/context, report the unresolvable state and the last error to the user, seeking further guidance.

8. Report Success: Once the command runs successfully and logs show no errors, report success to the user and present any relevant output or next steps (e.g., how to verify the environment).

Usage

Activate this skill when a shell command needs to be executed with automated error detection and self-correction.

Example: Use the auto-debug-command skill to run "docker-compose up --build"