Skip to main content
When a Bolt run completes, DinoAI automatically analyses the execution logs and generates a human-readable summary for each command. Instead of manually scanning through hundreds of log lines, you get a clear breakdown of what happened, what went wrong, and what to do next

​
What the Summary Includes

For each command executed in a Bolt run, DinoAI provides:
  • Execution overview — a plain-language summary of what the command did
  • Warnings and errors — surfaced and explained in context, not just raw stack traces
  • Recommendations — actionable suggestions to improve reliability and performance
Example: DinoAI Summary

Example: DinoAI Summary

​
Turbo CI summaries on GitHub

DinoAI summaries also appear on pull requests. When a Turbo CI run fails, Paradime posts the DinoAI Summary as a comment on the pull request — so reviewers can see why CI failed without leaving GitHub.
The PR comment includes the schedule name and commit SHA, a plain-language explanation of the failure, the specific errors with affected models, and the command that failed. A link back to the full run in Paradime is included for deeper investigation. No additional configuration is required — the DinoAI Summary is posted automatically whenever a Turbo CI run fails on a pull request. See the GitHub Turbo CI doc for details.

​
Troubleshooting Pipeline Runs

  1. Open the Bolt UI and select a schedule
  2. Scroll down to Run History and click on a specific run
  3. Scroll down to Logs and Artifacts
  4. Select a command that has an error or warning (e.g. dbt run)
  5. Click the Summary tab to see the DinoAI-generated analysis
Tip: The Summary tab is most useful when a command has failed or produced warnings — that’s when DinoAI’s analysis saves the most time.

​
Next Steps

Once you’ve reviewed the summary, you have two options:
  • Fix it yourself — use the recommendations to make changes in the IDE
  • Let DinoAI fix it — use Fix with DinoAI to trigger a one-click fix directly from the run

Related topics

Bolt Pipeline AgentSelf-HealingGitHubViewing Run Log HistorySchedules Dashboard