emergency-rescue
by gitgoodordietrying
Git & GitHubBrief
Recover from developer disasters.
Description
What This Skill Does
Step-by-step recovery procedures for common developer disasters: force-pushed branches, leaked credentials, full disks, failed database migrations, broken deploys, and lost SSH access. Each section follows a diagnose, fix, verify sequence. Commands are non-destructive by default, with destructive steps clearly flagged.
Having all recovery procedures in one place with a consistent diagnose-fix-verify structure means less time searching documentation during an active outage.
When to Use It
- Recovering lost commits after an accidental git reset --hard
- Revoking and purging a leaked API key from public git history
- Freeing disk space when Docker images exhaust a drive
- Rolling back a broken production deployment to a known-good state
- Regaining SSH access to a cloud server after a lockout
Example Workflow
User asks: I accidentally committed my AWS secret key to a public GitHub repo
- 1
Identify the file and commit containing the exposed credential using git log
- 2
Revoke the AWS access key immediately via the AWS IAM console before touching git
- 3
Remove the file from git tracking and add it to .gitignore
- 4
Rewrite git history with git filter-repo to permanently remove the secret from all commits
- 5
Force push the cleaned history to the remote and instruct all collaborators to re-clone
Credential revoked, git history cleaned of the secret, .gitignore updated to prevent recurrence, collaborators notified to re-clone
Installation
clawhub install gitgoodordietrying/emergency-rescueRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.