Improve failed-job recovery and in-product debugging

[dcrjodle]

Summary

When a job fails, users currently do not get enough help to diagnose the issue or recover without leaving the product. We need to improve the failed-job recovery experience with better debugging tools and guidance.

Acceptance Criteria

• Failed jobs expose useful logs and context
• Suggest likely causes for common failures
• Support targeted retry functionality
• Link to relevant help documentation from the failure state

Expected Behavior

Users should be able to:

1. View readable logs for failed jobs
2. Understand likely causes of the failure
3. Access remediation guidance
4. Retry jobs when appropriate
5. Access relevant help documentation without leaving the product

Proposed Solution

Add a debugging and recovery panel for failed jobs that includes:

• Readable logs and error context
• Likely causes for common failure scenarios
• Remediation guidance and next steps
• Step-by-step retry functionality when applicable

Additional Context

This finding came from the Tendril Test Submission.

Evidence

• Feedback Ref: RES-0004
• Test context: Windows, Claude, Stack: javascript (Node.js)
• Observed behavior: User did not see any feedback on why the CreatePlan job failed

Time index 10:18

Specific relevant feedback:

Did you get stuck during setup? Anything unclear?

Yes, I got stuck both at the setup stage and at the point of figuring out what to do once I create a plan inside the dashboard. Add images with guides, add a step-by-step guide, walk the person through it so they can see where your platform begins and where it ends. Show alerts when processes fail. As it was, I added a plan and it failed immediately, then silence. I have no idea whether a process started somewhere or not.

---

Original Slack thread

Reported by Christopher Clemons

[nielsbosma]

@phonofidelic Improve this please