When to use
- •Use for failed deploys, startup crashes, and intermittent production issues.
- •Use after deploys to validate stability.
Workflow
- •Pull recent logs for the active service/environment.
- •Identify first fatal error and repeated patterns.
- •Classify issue: config, dependency, runtime, network, DB, migration.
- •Propose the smallest safe fix and verification command.
Output expectations
- •Error signature (one line).
- •Suspected root cause.
- •Concrete next step to fix.
- •Verification step after fix.