When a migration fails, use the Diagnose Migration skill to classify the root cause and post a structured remediation plan as a comment on the issue so your team knows exactly what went wrong and what to do next.
Migration failures can be opaque. A comment on a backlog issue might say “migration failed” and link to a runner log with hundreds of lines. The agent cuts through that noise by reading the evidence and surfacing only what matters.
How to Use It
- Open the backlog issue in Migration HQ for the repository whose migration has failed.
- Assign the issue to the Packfiles Warp agent with the instructions: “Diagnose the migration failure.”
The agent categorizes the failure and posts a diagnosis comment directly on the issue.
What the Diagnosis Contains
The agent posts a comment on the issue with:
- Root cause — a single sentence stating what went wrong
- Remediation steps — a numbered list of specific actions to resolve the issue, including any Warp Vault changes, slash commands, or source system changes needed
- Retry instruction — the exact comment to post on the issue to retry the migration once remediation is complete
For most failure categories, the retry instruction is simply /migrate.
The too-big label (applied by Warp when a repository exceeds 10 GB) is always addressed in the diagnosis, even if another failure category is the primary cause. Size issues are a hard blocker that must be resolved regardless of what else is failing.
