Emergency Rescue Kit

**Judgement:** A strong, practical emergency runbook for developers and small ops teams. The content is technically competent, structured, and action-oriented, making it very useful as a first-response guide during common Git, credential, disk, process, and database incidents. **Key strengths** - Clear DIAGNOSE → FIX → VERIFY flow reduces the risk of making a bad situation worse. - Commands are mostly non-destructive, with dangerous operations explicitly flagged and often accompanied by safer alternatives. - Covers many of the highest-frequency, highest-stress problems individual developers encounter. - Includes both remediation and prevention patterns (revoke keys, add .gitignore, pre-commit hooks, transaction usage, connection limits). **Risks / limitations** - Some steps are inherently destructive (e.g., `git reset --hard`, `docker system prune -a`, log truncation, manual DB recovery) and could cause data loss if executed blindly or in the wrong environment. - The guidance is generic; it doesn’t account for project-specific constraints (e.g., managed DB backup setups, org-specific CI/CD, or compliance rules around credential rotation and log retention). - Page is truncated, so other areas (deploy rollbacks, SSL, SSH lockouts) may be present but are not fully assessable here. **Recommended scenarios** - As a go-to reference when a developer has just broken Git history, leaked credentials, or run out of disk on a dev/staging/CI machine. - As a training/mentoring aid for junior engineers learning how to recover safely from common operational mistakes. - As a starting template for teams to adapt into their own internal incident runbook, adding stack-specific and environment-specific safeguards. Use in production should still be accompanied by human judgment, awareness of backups and SLAs, and adaptation to local infrastructure and policies.