crash-triage

Crash Triage

Purpose

Handle crashes as high-severity defects with reproducibility, symbol quality, clustering, and release impact in mind.

Use When

• crashes are reported in QA or live builds
• stack traces are noisy
• the team needs root-cause prioritization

Inputs

• crash dumps or logs
• build symbols
• repro steps if any
• frequency and environment data

Process

1. group crashes by likely signature and root cause
2. separate deterministic repros from low-signal incidents
3. check if symbols and build metadata are sufficient
4. identify release blockers and likely owner systems
5. track verification once fixes land

Outputs

• crash cluster report
• owner assignment
• repro notes
• release impact summary

Quality Bar

• turns risk into explicit evidence and ownership
• keeps release blockers visible instead of implicit
• connects quality decisions to milestone and platform impact

Common Failure Modes

• severity inflation or minimization without player-impact context
• treating waived risks as invisible
• submission checklists that are incomplete or stale

Related Agents

• qa-lead
• build-engineer
• release-manager

Related Commands

• bug-triage
• release-check
• verify

Notes

• Keep this skill aligned with the relevant rules layer and current project documentation.
• If engine-specific constraints materially change the workflow, hand off to the matching engine skill or engine-specific reviewer.